lyzr-mcp
Provides tools for managing Git repositories, including pull, branch, commit, merge, pull request, and file operations.
Provides tools for training and querying knowledge graphs backed by Neo4j, including file, website, and text training and graph retrieval.
Provides tools for creating and managing WebRTC sessions for real-time communication.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@lyzr-mcp Create a knowledge base from our website and query it."
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
lyzr-mcp
An MCP server that exposes the Lyzr Enterprise API as MCP tools, resources, and prompts.
Bring-your-own-key, like the Supabase MCP server: every user supplies their own Lyzr API key. The server authenticates as that user and never uses a shared key.
Status: runnable. All three transports (stdio, SSE, Streamable HTTP), 423 tools across the agent + rag hosts — the original core set (agents, sessions, inference, RAG/knowledge base, parse/classify, credentials, live sources, knowledge graph, Cognis memory, scheduler, RAI) plus the full remaining
agent-devOpenAPI surface (A2A, agent eval, artifacts, assets, audit logs, channels, contexts, platform admin, GitAgent, human feedback, tool requests, skills, org LLM fallbacks, reports, semantic model, sessions v3, sharing, tools v3, traces, usage alerts, widget stream, user assets, world model, agent memory providers, ops, providers, workflows) plus the full remainingrag-devOpenAPI surface (Live Sources browse/webhooks/sync-permissions, Knowledge Graph Neo4j variants, Extract, doc content, Source Auth, file-based Parse/Train for PDF/DOCX/TXT/CSV/XLSX/PPTX/image, and the full KB Sync connector subsystem including deprecated legacy endpoints) — 5 resources, and 4 prompts. Endpoints confirmed against thelyzr-adkSDK and the Lyzr OpenAPI specs.
Documentation
📖 docs/README.md — architecture, setup, env vars, transports, HTTP contract, hosting, troubleshooting.
🧰 docs/tools.md — reference for the core tools, resources, and prompts.
Related MCP server: lynxprompt-mcp
Tools (363)
See .claude/skills/lyzr-api-endpoint/SKILL.md for the conventions
used to add these, and npx lyzr-mcp-skills below for end-user-facing skills covering the core groups.
Core groups
Group | Tools |
Agents & inference |
|
Sessions & templates |
|
Knowledge Base (RAG) |
|
Parse & classify |
|
Credentials & live sources |
|
Knowledge Graph (v4) |
|
Cognis memory |
|
Scheduler |
|
RAI guardrails |
|
Extended agent-dev groups (agent host)
Group | Tools |
Agent lifecycle (extra) |
|
Inference (extra) |
|
A2A agents |
|
Agent eval |
|
Artifacts |
|
Assets |
|
Audit logs |
|
Channels |
|
Contexts |
|
Platform admin (credits/flags/modules) |
|
GitAgent |
|
Human feedback / tool requests / skills |
|
Org LLM fallbacks |
|
Reports |
|
Semantic model |
|
Sessions v3 |
|
Sharing |
|
Tools v3 (core + integrations) |
|
Traces |
|
Usage alerts / widget stream / user assets |
|
World model |
|
Agent memory providers |
|
Ops |
|
Providers (core + credentials) |
|
Workflows |
|
Extended rag-dev groups (rag host)
Group | Tools |
Live sources (extra) |
|
Knowledge graph (Neo4j) |
|
Extract / doc content / source auth |
|
Parse (file uploads) |
|
Train (file uploads) |
|
KB Sync connectors |
|
KB Sync cc-pairs |
|
KB Sync legacy [deprecated] |
|
Resources (11): lyzr://agents, lyzr://agent/{agentId}, lyzr://kb/{ragId}, lyzr://kb/{ragId}/documents, lyzr://session/{sessionId}, lyzr://workflows, lyzr://workflow/{flowId}, lyzr://a2a-agents, lyzr://a2a-agent/{agentId}, lyzr://traces, lyzr://world-model/{agentId}.
Prompts (6): draft_support_agent, summarize_conversation, setup_rag_agent, draft_guardrail_policy, draft_a2a_agent, audit_agent_activity.
Skipped: KB Sync legacy OAuth/browse/webhooks are implemented but flagged
[deprecated]per the upstream API. Skills (client-side agent skill definitions) have no REST API and remain unwrapped.
How the API key flows
Transport | Key source |
stdio (local install) |
|
Streamable HTTP / SSE (hosted) | per-request header |
The server only reads the key — it never writes it to disk. The client (or the request header) is the secure store.
Build
npm install
npm run build # tsc -> build/index.js
npm test # vitest
npm run lint # biome lint .
npm run format # biome format --write .Requires Node ≥ 18 (uses the global fetch). Get your key from Agent Studio → Account → API Keys.
Add to your coding tool (local, stdio)
All examples run the built entrypoint: node <repo>/build/index.js stdio. Replace /ABS/PATH/lyzr-mcp with your clone's absolute path and YOUR_KEY with your Lyzr API key. (Once published to npm you can swap node <path> for npx -y lyzr-mcp.)
Claude Code
claude mcp add lyzr -e LYZR_API_KEY=YOUR_KEY -- node /ABS/PATH/lyzr-mcp/build/index.js stdioVerify with claude mcp list or /mcp. Add -s user for a global (all-projects) install.
Claude Desktop
~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": {
"lyzr": {
"command": "node",
"args": ["/ABS/PATH/lyzr-mcp/build/index.js", "stdio"],
"env": { "LYZR_API_KEY": "YOUR_KEY" }
}
}
}Fully quit and reopen Claude Desktop; the tools appear under the 🔨 icon.
OpenAI Codex CLI
~/.codex/config.toml:
[mcp_servers.lyzr]
command = "node"
args = ["/ABS/PATH/lyzr-mcp/build/index.js", "stdio"]
env = { LYZR_API_KEY = "YOUR_KEY" }(Or codex mcp add lyzr -- node /ABS/PATH/lyzr-mcp/build/index.js stdio, then set the key in the config's env.) List servers with codex mcp list.
Cursor
~/.cursor/mcp.json (global) or .cursor/mcp.json (per-project):
{
"mcpServers": {
"lyzr": {
"command": "node",
"args": ["/ABS/PATH/lyzr-mcp/build/index.js", "stdio"],
"env": { "LYZR_API_KEY": "YOUR_KEY" }
}
}
}Windsurf
~/.codeium/windsurf/mcp_config.json — same shape as Cursor (mcpServers → command/args/env).
VS Code (GitHub Copilot)
.vscode/mcp.json — note servers and type, and prompt for the key so it lands in the secret store:
{
"inputs": [
{ "type": "promptString", "id": "lyzr-key", "description": "Lyzr API Key", "password": true }
],
"servers": {
"lyzr": {
"type": "stdio",
"command": "node",
"args": ["/ABS/PATH/lyzr-mcp/build/index.js", "stdio"],
"env": { "LYZR_API_KEY": "${input:lyzr-key}" }
}
}
}Add to a hosted server (remote, bring-your-own-key)
Run one shared instance and let each user connect with their own key in a header:
PORT=3001 node /ABS/PATH/lyzr-mcp/build/index.js streamableHttp # endpoint: http://HOST:3001/mcpPut it behind HTTPS (reverse proxy, or ngrok http 3001 for a quick test). Then:
Claude Code:
claude mcp add --transport http lyzr https://your-host/mcp --header "x-api-key: YOUR_KEY"Cursor / Windsurf (mcp.json):
{ "mcpServers": { "lyzr": {
"url": "https://your-host/mcp",
"headers": { "x-api-key": "YOUR_KEY" } } } }VS Code (.vscode/mcp.json): { "servers": { "lyzr": { "type": "http", "url": "https://your-host/mcp", "headers": { "x-api-key": "YOUR_KEY" } } } }
stdio-only clients (Codex CLI, older clients) bridge to the remote server with mcp-remote:
# ~/.codex/config.toml
[mcp_servers.lyzr]
command = "npx"
args = ["-y", "mcp-remote", "https://your-host/mcp", "--header", "x-api-key:YOUR_KEY"]See docs/README.md for Docker + HTTPS deployment.
Skills
This repo ships nine Claude skills — short guides that teach Claude the right way to use this
server's tools (when to use lyzr_chat vs lyzr_stream_chat, how KB training works, when a destructive op
needs confirmation, etc). One skill per capability: lyzr-agents, lyzr-knowledge-base,
lyzr-knowledge-graph, lyzr-memory, lyzr-scheduler, lyzr-guardrails, lyzr-credentials,
lyzr-file-processing, lyzr-kb-sync.
There are two ways to install them, same as Supabase documents for its own agent-skills repo.
Option A: the community skills CLI (repo-agnostic)
Install all skills using the skills CLI, pointed at this
GitHub repo — no dependency on this package being published to npm, just a git checkout of
NeuralgoLyzr/lyzr-mcp:
npx skills add NeuralgoLyzr/lyzr-mcpInstall a specific skill:
npx skills add NeuralgoLyzr/lyzr-mcp --skill lyzr-agentsThis works because skills discovers any skills/<name>/SKILL.md folder with name/description
frontmatter in a repo — the exact layout under skills/ here — and copies the selected
skill(s) into .claude/skills/ (or the equivalent path for whichever coding agent you use).
Option B: the bundled lyzr-mcp-skills installer (zero GitHub dependency)
If you'd rather not depend on cloning this repo — e.g. air-gapped environments, or you just want the skills bundled with the npm package — use the installer shipped alongside this server:
npx lyzr-mcp-skills # interactive checklist
npx lyzr-mcp-skills all # install everything
npx lyzr-mcp-skills lyzr-agents lyzr-knowledge-base # install specific onesBoth install to .claude/skills/<name>/SKILL.md in your current directory.
Layout
src/
index.ts # launcher: argv → dynamic import of one transport
config.ts # per-user key helpers (env / header) + service-URL resolution
lyzr/ # one API client per Lyzr host, all extending LyzrHttp
http.ts # shared base: key, request(), query params, LyzrApiError, ServiceUrls
client.ts # agents + inference (agent host)
agent-extras.ts # sessions + templates (agent host)
rag.ts rag-admin.ts rag-content.ts rag-credentials.ts knowledge-graph.ts (rag host)
memory.ts scheduler.ts rai.ts
<28 more agent-host clients: a2a.ts, workflows.ts, world-model-*.ts, git-agent.ts, ...>
<8 more rag-host clients: kb-sync-*.ts, rag-parse-files.ts, rag-train-files.ts, ...>
server/ # createServer() factory, logging, roots
tools/ # one file per tool-set + index.ts registry (423 tools total)
resources/ # agents/KBs/sessions/workflows/A2A/traces/world-models as resources + subscriptions
prompts/ # reusable prompt templates
cli/ # install-skills.ts — the `lyzr-mcp-skills` installer
transports/ # stdio, sse (deprecated), streamableHttp
__tests__/ # vitest unit tests (one file per lyzr/*.ts client)
skills/ # end-user Claude skills, installed via `npx lyzr-mcp-skills`
.claude/skills/ # contributor-facing skill (lyzr-api-endpoint) for adding new endpointsAvailable Tools
423 toolslyzr_activate_agent_versionActivate Agent VersionA
Activate a specific version of an agent, making it the live version.
Inputs: agent_id (required), version_id (required).
lyzr_add_channel_agent_routeAdd Channel Agent RouteA
Add a new agent route to an existing channel.
Inputs: name (required), agent_id (required), channel_id (required).
lyzr_agent_chat_completionsAgent Chat CompletionsA
OpenAI-compatible chat/completions endpoint scoped to a specific agent.
Inputs: agent_id (required), messages (required), stream.
lyzr_agent_create_single_taskCreate Single-Task AgentC
Create a single-task template agent.
Inputs: name (required), model (required), top_p (required), provider_id (required), +7 more.
lyzr_agent_id_by_nameGet Agent Id by NameBRead-onlyIdempotent
Resolve an agent id by its name.
Inputs: agent_name (required).
lyzr_agent_published_sessionsList Agent Published SessionsBRead-onlyIdempotent
List published sessions for an agent.
Inputs: agent_id (required).
lyzr_agent_sessionsList Agent SessionsBRead-onlyIdempotent
List sessions for an agent.
Inputs: agent_id (required).
lyzr_agent_update_single_taskUpdate Single-Task AgentCIdempotent
Update a single-task template agent by id.
Inputs: name (required), model (required), top_p (required), agent_id (required), +8 more.
lyzr_bulk_delete_agentsBulk Delete AgentsADestructive
Permanently delete multiple agents by id.
Inputs: agent_ids (required).
lyzr_bulk_delete_tool_credentialsBulk Delete Tool CredentialsADestructive
Delete multiple tool credentials by id in one call.
Inputs: credential_ids (required).
lyzr_bulk_delete_toolsBulk Delete ToolsADestructive
Delete multiple tools by id in one call.
Inputs: tool_ids (required).
lyzr_bulk_delete_workflowsBulk Delete WorkflowsADestructive
Permanently delete multiple workflows by id.
Inputs: flow_ids (required).
lyzr_channel_webhookChannel WebhookA
Deliver an inbound messaging-platform webhook payload to a channel for processing.
Inputs: payload (required), channel_id (required).
lyzr_chatChat with Lyzr AgentB
Send a message to a Lyzr agent and get its response.
Inputs: message (required), agent_id (required), user_id, session_id.
lyzr_chat_completions_v4Chat Completions (v4)C
OpenAI-compatible v4 chat/completions endpoint (model + messages).
Inputs: model (required), messages (required), n, seed, +13 more.
lyzr_chat_with_fileChat With FileA
Send a chat message to an agent with an attached file (base64-encoded content).
Inputs: message (required), agent_id (required), session_id (required), file, +6 more.
lyzr_classifyClassify TextB
Classify text against a set of rules; returns the matching label(s).
Inputs: text (required), rules (required), extra_fields.
lyzr_clone_agentClone AgentA
Create a copy of an existing agent.
Inputs: agent_id (required).
lyzr_create_a2a_agentCreate A2A AgentA
Register a new A2A agent record pointing at a remote agent's base URL.
Inputs: base_url (required), name, skills, version, +11 more.
lyzr_create_aci_configurationCreate ACI ConfigurationB
Create an ACI app configuration for a given app id.
Inputs: app_id (required), security_scheme, security_scheme_overrides.
lyzr_create_aci_custom_appCreate ACI Custom AppC
Create a new ACI custom app with its functions.
Inputs: app_json (required), functions_json (required), secrets, skip_dry_run.
lyzr_create_agentCreate Lyzr AgentA
Create a new Lyzr agent. Returns the new agent_id to use with lyzr_chat.
Inputs: goal (required), name (required), role (required), instructions (required), +4 more.
lyzr_create_agent_evalCreate Agent EvalB
Create an agent evaluation config with a set of eval cases for an agent.
Inputs: agent_id (required), eval_name (required), session_id (required), agent_eval_list (required).
lyzr_create_agent_eval_resultCreate Agent Eval ResultA
Submit the results of running an agent evaluation (pass/fail outcomes per eval case).
Inputs: agent_id (required), agent_eval_id (required), agent_eval_result_list (required).
lyzr_create_artifactCreate ArtifactA
Create a new artifact for a user/session. Returns the created artifact.
Inputs: user_id (required), session_id (required), data, name, +3 more.
lyzr_create_bigquery_credentialCreate BigQuery Provider CredentialA
Create a BigQuery provider credential by uploading a service account JSON file.
Inputs: credential_data (required), service_account_json (required).
lyzr_create_channelCreate ChannelA
Create a channel configuration that routes an inbound messaging platform (e.g. telegram, slack) to Lyzr agents.
Inputs: config (required), platform (required), default_agent_id (required), agent_routes.
lyzr_create_composio_auth_configCreate Composio Auth ConfigC
Create a Composio auth config for a provider.
Inputs: client_id (required), provider_id (required), client_secret (required), scopes.
lyzr_create_contextCreate ContextA
Create a named key/value context. Returns the new context.
Inputs: name (required), value (required).
lyzr_create_feature_flagCreate Feature FlagC
Create a new feature flag.
Inputs: key (required), url (required), description (required), is_visible, +4 more.
lyzr_create_feedbackCreate Human FeedbackA
Submit human feedback on an agent's output for a given RAG feedback config.
Inputs: feedback (required), user_input (required), agent_output (required), feedback_rag_config_id (required).
lyzr_create_file_upload_credentialCreate File-Upload Provider CredentialA
Create a provider credential by uploading one or more files (e.g. certificates, keys).
Inputs: files (required), credential_data (required).
lyzr_create_inference_v4Create Inference (v4)B
Create a v4 inference response (OpenAI Responses-API style: model + input + tools).
Inputs: input (required), model (required), text, user, +12 more.
lyzr_create_lyzr_aci_toolCreate Lyzr ACI ToolC
Create a new org-wide Lyzr ACI tool.
Inputs: app_json (required), functions_json (required), form, secrets, +1 more.
lyzr_create_lyzr_providerCreate Lyzr ProviderC
Create a new Lyzr-managed provider connection.
Inputs: type (required), meta_data (required), provider_id (required).
lyzr_create_mcp_serverCreate MCP ServerB
Register a new MCP server (no-auth, api_key, or oauth).
Inputs: name (required), config (required), description, standard_server_id.
lyzr_create_moduleCreate ModuleB
Create a new sidebar module.
Inputs: key (required), url (required), description (required), beta, +21 more.
lyzr_create_oauth_tool_credentialCreate OAuth Tool CredentialC
Create an OAuth tool credential for a provider.
Inputs: user_id (required), provider_uuid (required), credential_name (required), scope, +7 more.
lyzr_create_providerCreate ProviderC
Create a new provider connection.
Inputs: type (required), meta_data (required), provider_id (required), form.
lyzr_create_provider_credentialCreate Provider CredentialA
Create a credential for a Lyzr provider (e.g. a database or SaaS integration). Returns the new credential_id.
Inputs: name (required), type (required), meta_data (required), credentials (required), +2 more.
lyzr_create_response_v4Create Response (v4)C
OpenAI-compatible v4 responses endpoint (model + input + tools).
Inputs: input (required), model (required), text, user, +12 more.
lyzr_create_static_tool_credentialCreate Static Tool CredentialC
Create a static (non-OAuth) tool credential.
Inputs: user_id (required), credentials (required), provider_uuid (required), credential_name (required).
lyzr_create_toolCreate ToolB
Create a new OpenAPI-based tool from an OpenAPI schema. Returns the created tool.
Inputs: tool_set_name (required), openapi_schema (required), user_id, openai_api_key, +5 more.
lyzr_create_tool_requestCreate Tool RequestA
Request a new tool, MCP server, skill, or custom integration be added to Lyzr.
Inputs: name (required), priority (required), use_case (required), description (required), +2 more.
lyzr_create_webrtc_sessionCreate WebRTC Voice SessionA
Create a WebRTC voice session for an agent + voice id.
Inputs: agent_id (required), voice_id (required).
lyzr_create_workflowCreate WorkflowC
Create a new workflow.
Inputs: flow_name (required), api_key, flow_data.
lyzr_credential_createCreate CredentialB
Create a new credential. Returns the created credential.
Inputs: name (required), credentials (required), provider_id (required), type, +3 more.
lyzr_credential_deleteDelete CredentialADestructiveIdempotent
Permanently delete a credential by id.
Inputs: credential_id (required).
lyzr_credential_getGet CredentialARead-onlyIdempotent
Fetch a credential by id.
Inputs: credential_id (required).
lyzr_credential_listList CredentialsBRead-onlyIdempotent
List all credentials.
Inputs: no named parameters.
lyzr_credential_updateUpdate CredentialAIdempotent
Update an existing credential. All fields optional.
Inputs: credential_id (required), name, scope, metadata, +2 more.
lyzr_delete_a2a_agentDelete A2A AgentADestructiveIdempotent
Permanently delete an A2A agent record by id.
Inputs: agent_id (required).
lyzr_delete_aci_configurationDelete ACI ConfigurationADestructiveIdempotent
Delete an ACI app configuration by app id.
Inputs: app_id (required).
lyzr_delete_aci_connectionDelete ACI ConnectionADestructiveIdempotent
Delete an ACI linked account connection by its id.
Inputs: linked_account_id (required).
lyzr_delete_aci_custom_appDelete ACI Custom AppADestructiveIdempotent
Permanently delete an ACI custom app by id.
Inputs: app_id (required).
lyzr_delete_agentDelete Lyzr AgentADestructiveIdempotent
Permanently delete a Lyzr agent by its agent_id.
Inputs: agent_id (required).
lyzr_delete_artifactDelete ArtifactADestructiveIdempotent
Permanently delete an artifact by id, scoped to a user and session.
Inputs: user_id (required), session_id (required), artifact_id (required).
lyzr_delete_assetDelete AssetADestructiveIdempotent
Permanently delete an asset by id.
Inputs: asset_id (required).
lyzr_delete_channelDelete ChannelADestructiveIdempotent
Permanently delete a channel configuration by id.
Inputs: channel_id (required).
lyzr_delete_composio_auth_configDelete Composio Auth ConfigBDestructiveIdempotent
Delete the Composio auth config for a provider.
Inputs: provider_id (required).
lyzr_delete_composio_connectionDelete Composio ConnectionADestructiveIdempotent
Delete a Composio connection by id.
Inputs: connection_id (required).
lyzr_delete_contextDelete ContextADestructiveIdempotent
Permanently delete a context by id.
Inputs: context_id (required).
lyzr_delete_feature_flagDelete Feature FlagBDestructiveIdempotent
Permanently delete a feature flag by key.
Inputs: key (required).
lyzr_delete_lyzr_aci_toolDelete Lyzr ACI ToolADestructiveIdempotent
Permanently delete a Lyzr ACI tool by provider id.
Inputs: provider_id (required).
lyzr_delete_mcp_serverDelete MCP ServerADestructiveIdempotent
Delete a registered MCP server.
Inputs: server_id (required).
lyzr_delete_moduleDelete ModuleADestructiveIdempotent
Permanently delete a sidebar module by key.
Inputs: key (required).
lyzr_delete_providerDelete ProviderADestructiveIdempotent
Permanently delete a provider by id.
Inputs: provider_id (required).
lyzr_delete_provider_credentialDelete Provider CredentialADestructiveIdempotent
Permanently delete a provider credential by id.
Inputs: credential_id (required).
lyzr_delete_toolDelete ToolADestructiveIdempotent
Delete a tool by id.
Inputs: tool_id (required).
lyzr_delete_tool_credentialDelete Tool CredentialBDestructiveIdempotent
Delete a tool credential by id.
Inputs: credential_id (required).
lyzr_delete_workflowDelete WorkflowADestructiveIdempotent
Permanently delete a workflow by id.
Inputs: flow_id (required).
lyzr_execute_mcp_toolExecute MCP ToolB
Invoke a tool exposed by a registered MCP server.
Inputs: server_id (required), tool_name (required), arguments.
lyzr_execute_toolExecute ToolA
Execute a named tool (with optional tool configs) directly, without a full chat turn.
Inputs: tool_name (required), agent_id, trace_id, arguments, +1 more.
lyzr_execute_workflowExecute WorkflowC
Execute a workflow, optionally passing input data.
Inputs: flow_id (required), input_data.
lyzr_extractExtract Structured DataB
Extract structured data from a file (base64), a file URL, or raw text, according to an extraction schema.
Inputs: extraction_schema (required), tier, target, annotate, +9 more.
lyzr_get_a2a_agentGet A2A AgentARead-onlyIdempotent
Fetch a registered A2A agent record by id.
Inputs: agent_id (required).
lyzr_get_a2a_agent_cardGet A2A Agent CardARead-onlyIdempotent
Fetch the A2A agent card (the .well-known/agent-card.json descriptor) for a served agent.
Inputs: agent_id (required).
lyzr_get_a2a_agent_card_convenienceGet A2A Agent Card (Convenience)ARead-onlyIdempotent
Fetch the A2A agent card via the convenience path (without .well-known/agent-card.json).
Inputs: agent_id (required).
lyzr_get_activity_metricsGet Activity MetricsARead-onlyIdempotent
Get combined DAU/MAU activity metrics for a given date (defaults to today UTC).
Inputs: date.
lyzr_get_agentGet Lyzr AgentARead-onlyIdempotent
Fetch the details of a single Lyzr agent by its agent_id.
Inputs: agent_id (required).
lyzr_get_agent_evalGet Agent EvalARead-onlyIdempotent
Get agent evaluation configs for a given agent.
Inputs: agent_id (required).
lyzr_get_agent_eval_resultGet Agent Eval ResultCRead-onlyIdempotent
Get agent evaluation results by agent eval id.
Inputs: agent_eval_id (required).
lyzr_get_agent_eval_result_by_agentGet Agent Eval Result By AgentBRead-onlyIdempotent
Get agent evaluation results for a given agent id.
Inputs: agent_id (required).
lyzr_get_agent_stale_connectionsGet Agent Stale ConnectionsARead-onlyIdempotent
Get an agent's stale tool connections (connections that need reauthorization).
Inputs: user_id (required), agent_id (required).
lyzr_get_agent_versionGet Agent VersionBRead-onlyIdempotent
Fetch a specific version of an agent.
Inputs: agent_id (required), version_id (required).
lyzr_get_all_toolsGet All ToolsBRead-onlyIdempotent
List all tools available across providers.
Inputs: no named parameters.
lyzr_get_artifactGet ArtifactARead-onlyIdempotent
Fetch a single artifact by id, scoped to a user and session.
Inputs: user_id (required), session_id (required), artifact_id (required).
lyzr_get_assetGet AssetARead-onlyIdempotent
Fetch asset metadata by id.
Inputs: asset_id (required).
lyzr_get_asset_parse_statusGet Asset Parse StatusARead-onlyIdempotent
Get the parsing status of an asset by id.
Inputs: asset_id (required).
lyzr_get_asset_rawGet Asset Raw ContentARead-onlyIdempotent
Fetch the raw content of an asset by id.
Inputs: asset_id (required).
lyzr_get_audit_logGet Audit Log By IDARead-onlyIdempotent
Fetch a single audit log entry by its id.
Inputs: log_id (required).
lyzr_get_audit_statsGet Audit StatsARead-onlyIdempotent
Get aggregated audit log statistics for an optional time range.
Inputs: end_time, start_time.
lyzr_get_cached_creditsGet Cached CreditsARead-onlyIdempotent
Fetch the cached credit-cost map keyed by resource/action.
Inputs: x_server_token.
lyzr_get_chat_task_statusGet Chat Task StatusARead-onlyIdempotent
Poll the status/result of a chat task submitted via lyzr_submit_chat_task.
Inputs: task_id (required).
lyzr_get_composio_action_limitGet Composio Action LimitARead-onlyIdempotent
Get the current Composio action usage/limit.
Inputs: no named parameters.
lyzr_get_composio_auth_configGet Composio Auth ConfigARead-onlyIdempotent
Fetch the Composio auth config for a provider.
Inputs: provider_id (required).
lyzr_get_connected_accountsGet Connected AccountsBRead-onlyIdempotent
List a user's connected tool accounts/credentials.
Inputs: user_id (required).
lyzr_get_contextGet ContextARead-onlyIdempotent
Fetch a context by id.
Inputs: context_id (required).
lyzr_get_context_by_name_internalGet Context By Name (Internal)ARead-onlyIdempotent
Internal endpoint: fetch a context by its name using an internal api_key.
Inputs: api_key (required), context_name (required).
lyzr_get_contexts_countGet Contexts CountBRead-onlyIdempotent
Get the total count of contexts.
Inputs: no named parameters.
lyzr_get_context_usageGet Context UsageARead-onlyIdempotent
Get usage information for a context by id.
Inputs: context_id (required).
lyzr_get_context_value_internalGet Context Value (Internal)BRead-onlyIdempotent
Internal endpoint: fetch a context's value by its name using an internal api_key.
Inputs: api_key (required), context_name (required).
lyzr_get_credential_agentsGet Credential AgentsARead-onlyIdempotent
List agents that use a given tool credential.
Inputs: credential_id (required), provider_uuid.
lyzr_get_dauGet Daily Active UsersARead-onlyIdempotent
Get the daily active user count for a given date (defaults to today UTC).
Inputs: date.
lyzr_get_dau_trendGet DAU TrendARead-onlyIdempotent
Get the daily active users trend over the last N days.
Inputs: days.
lyzr_get_feature_flag_adminGet Feature Flag (Admin)ARead-onlyIdempotent
Fetch a single feature flag by key.
Inputs: key (required).
lyzr_get_feature_flagsGet Feature FlagsARead-onlyIdempotent
Get the resolved feature flags applicable to the current caller.
Inputs: no named parameters.
lyzr_get_featuresGet FeaturesARead-onlyIdempotent
Get the aggregate feature-availability map for the current caller.
Inputs: no named parameters.
lyzr_get_mauGet Monthly Active UsersARead-onlyIdempotent
Get the monthly active user count for the month containing a given date (defaults to current month UTC).
Inputs: date.
lyzr_get_mau_trendGet MAU TrendARead-onlyIdempotent
Get the monthly active users trend over the last N months.
Inputs: months.
lyzr_get_mcp_oauth_statusGet MCP OAuth StatusARead-onlyIdempotent
Check the status of an in-progress MCP OAuth flow.
Inputs: state (required), server_id (required).
lyzr_get_mcp_server_agentsGet MCP Server AgentsARead-onlyIdempotent
List agents that use a given MCP server.
Inputs: server_id (required).
lyzr_get_module_adminGet Module (Admin)ARead-onlyIdempotent
Fetch a single sidebar module by key.
Inputs: key (required).
lyzr_get_modulesGet ModulesARead-onlyIdempotent
Get the resolved sidebar modules (nav + footer) for the current caller.
Inputs: no named parameters.
lyzr_get_multiple_context_values_internalGet Multiple Context Values (Internal)B
Internal endpoint: fetch multiple context values by name using an internal api_key.
Inputs: api_key (required), context_names (required).
lyzr_get_org_llm_fallbacksGet Org LLM FallbacksARead-onlyIdempotent
Get the organization's configured LLM fallback chain.
Inputs: no named parameters.
lyzr_get_providerGet ProviderBRead-onlyIdempotent
Fetch a provider by id.
Inputs: provider_id (required).
lyzr_get_provider_credentialGet Provider CredentialBRead-onlyIdempotent
Fetch a provider credential by id.
Inputs: credential_id (required).
lyzr_get_providers_by_typeGet Providers By TypeARead-onlyIdempotent
List providers of a given type, paginated.
Inputs: provider_type (required), page, limit.
lyzr_get_skill_usageGet Skill UsageARead-onlyIdempotent
Get usage statistics for a given skill.
Inputs: skill_id (required).
lyzr_get_task_statusGet Lyzr Task StatusARead-onlyIdempotent
Check the status and (when complete) result of a long-running Lyzr task.
Inputs: task_id (required).
lyzr_get_toolGet ToolARead-onlyIdempotent
Fetch a tool by id.
Inputs: tool_id (required).
lyzr_get_tools_actionsGet Tools ActionsBRead-onlyIdempotent
List available tool actions for a provider.
Inputs: provider_identifier (required), app_id, tool_source.
lyzr_get_trace_detailsGet Trace DetailsARead-onlyIdempotent
Fetch raw details for a trace by id.
Inputs: trace_id (required).
lyzr_get_trace_ganttGet Trace Gantt ChartBRead-onlyIdempotent
Fetch the Gantt chart (span tree with timings) for a trace.
Inputs: trace_id (required).
lyzr_get_traces_dashboardGet Traces Dashboard MetricsARead-onlyIdempotent
Fetch aggregate dashboard metrics (credits, traces, spans, tokens, latency, error rate) across traces, optionally filtered.
Inputs: agent_id, end_time, session_id, start_time, +2 more.
lyzr_get_trace_summaryGet Trace SummaryARead-onlyIdempotent
Fetch the detailed summary (agent, LLM, tool call, token, and cost info) for a trace.
Inputs: trace_id (required).
lyzr_get_workflowGet WorkflowARead-onlyIdempotent
Fetch a workflow by id.
Inputs: flow_id (required).
lyzr_git_create_branchCreate BranchA
Create a new branch in an agent's git repo.
Inputs: agent_id (required), branch_name (required), from_branch.
lyzr_git_create_prCreate PRB
Create a pull request between two branches of an agent's repo.
Inputs: title (required), agent_id (required), source_branch (required), target_branch (required), +2 more.
lyzr_git_delete_governanceDelete GovernanceADestructiveIdempotent
Delete the governance policy for an agent's git repo.
Inputs: agent_id (required).
lyzr_git_deploy_branchDeploy BranchC
Deploy a branch of an agent's git repo.
Inputs: branch (required), agent_id (required).
lyzr_git_disconnectDisconnect GitADestructiveIdempotent
Disconnect the git integration for an agent.
Inputs: agent_id (required), purge.
lyzr_git_get_commit_snapshotGet Commit SnapshotBRead-onlyIdempotent
Get a snapshot of the repo state at a specific commit.
Inputs: agent_id (required), commit_sha (required).
lyzr_git_get_file_contentGet File ContentARead-onlyIdempotent
Get the content of a file in an agent's git repo.
Inputs: agent_id (required), file_path (required), branch.
lyzr_git_get_governanceGet GovernanceARead-onlyIdempotent
Get the governance policy for an agent's git repo.
Inputs: agent_id (required).
lyzr_git_get_statusGet Git StatusBRead-onlyIdempotent
Get the current git status for an agent's repo.
Inputs: agent_id (required).
lyzr_git_init_repoInit Git RepoC
Initialize the git repository for an agent.
Inputs: agent_id (required).
lyzr_git_list_branchesList BranchesARead-onlyIdempotent
List branches in an agent's git repo.
Inputs: agent_id (required).
lyzr_git_list_commitsList CommitsARead-onlyIdempotent
List commits for an agent's git repo, optionally filtered by branch.
Inputs: agent_id (required), branch.
lyzr_git_list_prsList PRsARead-onlyIdempotent
List pull requests for an agent's git repo.
Inputs: agent_id (required).
lyzr_git_list_repo_filesList Repo FilesARead-onlyIdempotent
List files in an agent's git repo, optionally at a specific branch.
Inputs: agent_id (required), branch.
lyzr_git_merge_branchesMerge BranchesB
Merge one branch into another in an agent's git repo.
Inputs: agent_id (required), source_branch (required), target_branch (required), commit_message.
lyzr_git_pullPull From GitA
Pull the latest changes from git for an agent's repo.
Inputs: agent_id (required), branch.
lyzr_git_restore_commitRestore CommitADestructive
Restore the repo to the state of a specific commit.
Inputs: agent_id (required), commit_sha (required).
lyzr_git_save_configSave Git ConfigA
Save or update the git connection (provider, repo, credentials) for an agent.
Inputs: agent_id (required), repo_name (required), org, pat, +4 more.
lyzr_git_save_file_contentSave File ContentA
Save (create or update) a file's content in an agent's git repo.
Inputs: content (required), agent_id (required), file_path (required), branch, +1 more.
lyzr_git_save_governanceSave GovernanceA
Save or update the governance (approval pipeline) policy for an agent's git repo.
Inputs: agent_id (required), name, gates, entity_type.
lyzr_git_switch_branchSwitch BranchA
Switch the active branch for an agent's git repo.
Inputs: branch (required), agent_id (required).
lyzr_git_sync_governanceSync GovernanceC
Sync the governance policy for an agent's git repo.
Inputs: agent_id (required).
lyzr_git_update_reviewersUpdate ReviewersA
Update the list of reviewers for an agent's git repo.
Inputs: agent_id (required), reviewers (required).
lyzr_git_validate_configValidate Git ConfigA
Validate a git connection configuration for an agent without saving it.
Inputs: agent_id (required), repo_name (required), org, pat, +4 more.
lyzr_infer_a2a_agentInfer A2A AgentB
Send a message to a registered A2A agent and get its response.
Inputs: message (required), agent_id (required), context_id.
lyzr_initiate_mcp_oauthInitiate MCP OAuth FlowA
Start the OAuth flow for an MCP server; returns an auth URL.
Inputs: server_id (required).
lyzr_is_tool_credential_test_supportedIs Tool Credential Test SupportedBRead-onlyIdempotent
Check whether a tool credential supports test invocation.
Inputs: credential_id (required).
lyzr_kb_bulk_deleteBulk Delete Knowledge BasesB
Permanently delete multiple knowledge base configs by id.
Inputs: config_ids (required).
lyzr_kb_createCreate Knowledge BaseA
Create a RAG knowledge base. Returns its id.
Inputs: name (required), llm_model, description, vector_store, +1 more.
lyzr_kb_deleteDelete Knowledge BaseADestructiveIdempotent
Permanently delete a knowledge base by id.
Inputs: kb_id (required).
lyzr_kb_delete_docsDelete KB DocumentsADestructiveIdempotent
Delete documents from a knowledge base by their document ids.
Inputs: docs (required), rag_id (required).
lyzr_kb_delete_docs_by_filterDelete KB Documents by FilterADestructiveIdempotent
Delete documents from a knowledge base matching a filter and/or a list of document ids.
Inputs: rag_id (required), docs, filters, extra_fields.
lyzr_kb_getGet Knowledge BaseARead-onlyIdempotent
Fetch a knowledge base by id.
Inputs: kb_id (required).
lyzr_kb_listList Knowledge BasesBRead-onlyIdempotent
List knowledge bases for a user.
Inputs: user_id (required).
lyzr_kb_list_documentsList KB DocumentsBRead-onlyIdempotent
List the documents indexed in a knowledge base.
Inputs: rag_id (required).
lyzr_kb_queryQuery Knowledge BaseBRead-onlyIdempotent
Retrieve relevant chunks from a knowledge base.
Inputs: query (required), rag_id (required), top_k.
lyzr_kb_resetReset Knowledge BaseBDestructiveIdempotent
Clear all documents from a knowledge base.
Inputs: rag_id (required).
lyzr_kb_sync_browse_childrenBrowse SharePoint Drive Children (Deprecated)ARead-onlyIdempotent
[deprecated] Browse files/folders within a SharePoint drive path.
Inputs: site_url (required), drive_name (required), credential_id (required), folder_path.
lyzr_kb_sync_browse_drivesBrowse SharePoint Drives (Deprecated)BRead-onlyIdempotent
[deprecated] Browse drives within a SharePoint site.
Inputs: site_url (required), credential_id (required).
lyzr_kb_sync_browse_sitesBrowse SharePoint Sites (Deprecated)BRead-onlyIdempotent
[deprecated] Browse available SharePoint sites.
Inputs: credential_id (required).
lyzr_kb_sync_cc_pairs_cancel_attemptCancel KB Sync Index AttemptADestructiveIdempotent
Cancel an in-progress index attempt.
Inputs: attempt_id (required), cc_pair_id (required).
lyzr_kb_sync_cc_pairs_createCreate KB Sync Connector-Credential PairA
Create a connector-credential pair, linking a connector to a credential (and optionally a knowledge base).
Inputs: connector_id (required), credential_id (required), name, rag_id.
lyzr_kb_sync_cc_pairs_getGet KB Sync Connector-Credential PairARead-onlyIdempotent
Fetch a connector-credential pair by id.
Inputs: cc_pair_id (required).
lyzr_kb_sync_cc_pairs_get_attemptGet KB Sync Index AttemptARead-onlyIdempotent
Fetch a single index attempt by id.
Inputs: attempt_id (required), cc_pair_id (required).
lyzr_kb_sync_cc_pairs_listList KB Sync Connector-Credential PairsARead-onlyIdempotent
List all connector-credential pairs.
Inputs: no named parameters.
lyzr_kb_sync_cc_pairs_list_attemptsList KB Sync Index AttemptsBRead-onlyIdempotent
List index attempts (sync runs) for a connector-credential pair.
Inputs: cc_pair_id (required), limit, offset.
lyzr_kb_sync_cc_pairs_pausePause KB Sync Connector-Credential PairAIdempotent
Pause syncing for a connector-credential pair.
Inputs: cc_pair_id (required).
lyzr_kb_sync_cc_pairs_resumeResume KB Sync Connector-Credential PairCIdempotent
Resume syncing for a connector-credential pair.
Inputs: cc_pair_id (required).
lyzr_kb_sync_cc_pairs_statusGet KB Sync StatusBRead-onlyIdempotent
Get the current sync status for a connector-credential pair, including its active and last completed index attempts.
Inputs: cc_pair_id (required).
lyzr_kb_sync_cc_pairs_syncTrigger KB SyncB
Trigger a sync run for a connector-credential pair.
Inputs: cc_pair_id (required).
lyzr_kbsync_connector_createCreate KB Sync ConnectorB
Create a new KB Sync connector for a document source.
Inputs: name (required), source (required), connector_specific_config (required).
lyzr_kbsync_connector_deleteDelete KB Sync ConnectorADestructiveIdempotent
Permanently delete a KB Sync connector by id.
Inputs: connector_id (required).
lyzr_kbsync_connector_getGet KB Sync ConnectorARead-onlyIdempotent
Fetch a KB Sync connector by id.
Inputs: connector_id (required).
lyzr_kbsync_connector_listList KB Sync ConnectorsARead-onlyIdempotent
List all KB Sync connectors.
Inputs: no named parameters.
lyzr_kbsync_connector_updateUpdate KB Sync ConnectorAIdempotent
Update a KB Sync connector's name, config, or disabled state.
Inputs: connector_id (required), name, disabled, connector_specific_config.
lyzr_kbsync_credential_createCreate KB Sync CredentialA
Create a KB Sync connector credential (distinct from the general RAG credentials system).
Inputs: credential_json (required), scope.
lyzr_kbsync_credential_listList KB Sync CredentialsARead-onlyIdempotent
List KB Sync connector credentials (distinct from the general RAG credentials system).
Inputs: no named parameters.
lyzr_kb_sync_validate_site_accessValidate SharePoint Site Access (Deprecated)BIdempotent
[deprecated] Validate that a credential has access to the given SharePoint site/drive URLs.
Inputs: site_urls (required), credential_id (required).
lyzr_kb_sync_webhook_notificationKB Sync Webhook Notification (Deprecated)C
[deprecated] Receive a KB Sync webhook notification payload.
Inputs: payload.
lyzr_kb_sync_webhook_validationKB Sync Webhook Validation (Deprecated)BRead-onlyIdempotent
[deprecated] Webhook validation handshake endpoint.
Inputs: no named parameters.
lyzr_kb_train_documentsTrain KB DocumentsB
Ingest an array of document objects into a knowledge base. Each document is a free-form object.
Inputs: rag_id (required), documents (required).
lyzr_kb_train_docxTrain KB from DOCXB
Upload and train a DOCX file into a knowledge base.
Inputs: rag_id (required), filename (required), file_content_base64 (required), mime_type, +5 more.
lyzr_kb_train_imageTrain KB from ImageB
Upload and train an image file into a knowledge base.
Inputs: rag_id (required), filename (required), file_content_base64 (required), mime_type, +4 more.
lyzr_kb_train_pdfTrain KB from PDFC
Upload and train a PDF file into a knowledge base.
Inputs: rag_id (required), filename (required), file_content_base64 (required), mime_type, +5 more.
lyzr_kb_train_pptxTrain KB from PPTXB
Upload and train a PPTX presentation into a knowledge base.
Inputs: rag_id (required), filename (required), file_content_base64 (required), mime_type, +4 more.
lyzr_kb_train_textTrain KB from TextC
Ingest text chunks into a knowledge base.
Inputs: texts (required), rag_id (required).
lyzr_kb_train_txt_fileTrain KB from Text FileB
Upload and train a plain-text (.txt) file into a knowledge base.
Inputs: rag_id (required), filename (required), file_content_base64 (required), mime_type, +4 more.
lyzr_kb_train_websiteTrain KB from WebsiteA
Crawl and ingest one or more URLs into a knowledge base.
Inputs: urls (required), rag_id (required).
lyzr_kb_train_xlsxTrain KB from XLSXA
Upload and train an XLSX spreadsheet into a knowledge base.
Inputs: rag_id (required), filename (required), file_content_base64 (required), mime_type, +4 more.
lyzr_kb_updateUpdate Knowledge Base ConfigAIdempotent
Update a knowledge base config by id. Provide the fields to change via extra_fields.
Inputs: config_id (required), extra_fields.
lyzr_kb_update_docs_metadataUpdate KB Documents MetadataAIdempotent
Update metadata on documents in a knowledge base matching a filter.
Inputs: rag_id (required), filters, extra_fields, metadata_updates.
lyzr_kg_deduplicateDeduplicate Knowledge GraphC
Deduplicate entities in a knowledge base's graph.
Inputs: rag_id (required).
lyzr_kg_ext_deduplicate_neo4jDeduplicate Neo4j Knowledge Graph EntitiesBIdempotent
Deduplicate entities in the Neo4j knowledge graph for a RAG system (namespaced endpoint).
Inputs: rag_id (required).
lyzr_kg_ext_get_graph_neo4jGet Neo4j Knowledge GraphARead-onlyIdempotent
Fetch the Neo4j knowledge graph nodes for a RAG system (namespaced endpoint).
Inputs: rag_id (required), limit.
lyzr_kg_ext_train_fileTrain Knowledge Graph From FileC
Upload a file (base64-encoded) to train the Neo4j knowledge graph.
Inputs: rag_id (required), filename (required), file_content_base64 (required), mime_type, +2 more.
lyzr_kg_ext_train_file_neo4jTrain Neo4j Knowledge Graph From FileC
Upload a file (base64-encoded) to train the Neo4j knowledge graph (namespaced endpoint).
Inputs: rag_id (required), filename (required), file_content_base64 (required), mime_type, +2 more.
lyzr_kg_ext_train_file_taskTrain Knowledge Graph From File (Async Task)C
Upload a file (base64-encoded) to train the Neo4j knowledge graph as an async task.
Inputs: rag_id (required), filename (required), file_content_base64 (required), mime_type, +2 more.
lyzr_kg_ext_train_file_task_neo4jTrain Neo4j Knowledge Graph From File (Async Task)A
Upload a file (base64-encoded) to train the Neo4j knowledge graph as an async task (namespaced endpoint).
Inputs: rag_id (required), filename (required), file_content_base64 (required), mime_type, +2 more.
lyzr_kg_ext_train_text_neo4jTrain Neo4j Knowledge Graph From TextB
Ingest text into the Neo4j knowledge graph (namespaced endpoint).
Inputs: text (required), rag_id (required), source (required), extra_info, +2 more.
lyzr_kg_ext_train_text_task_neo4jTrain Neo4j Knowledge Graph From Text (Async Task)B
Ingest text into the Neo4j knowledge graph as an async task (namespaced endpoint).
Inputs: text (required), rag_id (required), source (required), extra_info, +2 more.
lyzr_kg_ext_train_website_neo4jTrain Neo4j Knowledge Graph From WebsiteC
Crawl and ingest websites into the Neo4j knowledge graph (namespaced endpoint).
Inputs: urls (required), rag_id (required), actor, source, +7 more.
lyzr_kg_get_graphGet Knowledge GraphBRead-onlyIdempotent
Fetch the knowledge graph for a knowledge base.
Inputs: rag_id (required), limit.
lyzr_kg_task_statusGet Knowledge Graph Task StatusARead-onlyIdempotent
Poll the status of an async knowledge graph task.
Inputs: task_id (required).
lyzr_kg_train_textTrain Knowledge Graph from TextC
Ingest text into the v4 knowledge graph.
Inputs: text (required), rag_id (required), source (required), extra_info, +2 more.
lyzr_kg_train_text_taskTrain Knowledge Graph from Text (Async)A
Ingest text into the v4 knowledge graph as an async task; returns a task id to poll.
Inputs: text (required), rag_id (required), source (required), extra_info, +2 more.
lyzr_kg_train_websiteTrain Knowledge Graph from WebsiteA
Crawl and ingest one or more URLs into the v4 knowledge graph.
Inputs: urls (required), rag_id (required), actor, source, +7 more.
lyzr_kill_switch_traceKill Switch Trace AgentsADestructive
Trigger the kill switch to stop agents running under a trace.
Inputs: trace_id (required).
lyzr_list_a2a_agentsList A2A AgentsARead-onlyIdempotent
List all registered A2A (agent-to-agent) agents.
Inputs: no named parameters.
lyzr_list_aci_configurationsList ACI ConfigurationsARead-onlyIdempotent
List ACI (Aipolabs Composio-alternative) app configurations.
Inputs: no named parameters.
lyzr_list_agentsList Lyzr AgentsBRead-onlyIdempotent
List the agents available to your Lyzr API key.
Inputs: no named parameters.
lyzr_list_agent_versionsList Agent VersionsARead-onlyIdempotent
List the version history of an agent.
Inputs: agent_id (required).
lyzr_list_all_channelsList All ChannelsBRead-onlyIdempotent
List every channel configuration.
Inputs: no named parameters.
lyzr_list_all_user_toolsList All User ToolsARead-onlyIdempotent
List all tools available to the current user.
Inputs: no named parameters.
lyzr_list_artifactsList ArtifactsBRead-onlyIdempotent
List artifacts with optional pagination and filters.
Inputs: page, limit, user_id, session_id, +1 more.
lyzr_list_artifacts_by_sessionList Artifacts by User SessionARead-onlyIdempotent
List paginated artifacts belonging to a specific user's session.
Inputs: user_id (required), session_id (required), page, limit.
lyzr_list_assetsList AssetsCRead-onlyIdempotent
List assets, paginated.
Inputs: page, limit.
lyzr_list_channelsList ChannelsARead-onlyIdempotent
List channel configurations for a given agent.
Inputs: agent_id (required).
lyzr_list_contextsList ContextsARead-onlyIdempotent
List contexts, with pagination.
Inputs: skip, limit.
lyzr_list_feature_flags_adminList Feature Flags (Admin)ARead-onlyIdempotent
List all feature flags with their gating configuration.
Inputs: no named parameters.
lyzr_list_lyzr_aci_toolsList Lyzr ACI ToolsARead-onlyIdempotent
List org-wide Lyzr ACI tools.
Inputs: no named parameters.
lyzr_list_mcp_server_promptsList MCP Server PromptsARead-onlyIdempotent
List the prompts exposed by an MCP server.
Inputs: server_id (required).
lyzr_list_mcp_server_resourcesList MCP Server ResourcesARead-onlyIdempotent
List the resources exposed by an MCP server.
Inputs: server_id (required).
lyzr_list_mcp_serversList MCP ServersARead-onlyIdempotent
List MCP servers registered for the account.
Inputs: no named parameters.
lyzr_list_mcp_server_toolsList MCP Server ToolsARead-onlyIdempotent
List the tools exposed by an MCP server.
Inputs: server_id (required).
lyzr_list_modules_adminList Modules (Admin)ARead-onlyIdempotent
List all sidebar modules with their configuration.
Inputs: no named parameters.
lyzr_list_my_audit_logsList My Audit LogsBRead-onlyIdempotent
Get audit logs for the current authenticated user, with optional filters.
Inputs: limit, action, offset, result, +4 more.
lyzr_list_org_agentsList Org AgentsARead-onlyIdempotent
List agents across the organization, with optional search and pagination.
Inputs: page, limit, search.
lyzr_list_org_audit_logsList Organization Audit LogsBRead-onlyIdempotent
Get audit logs for the whole organization, with optional filters.
Inputs: limit, action, offset, result, +7 more.
lyzr_list_provider_credentials_by_typeList Provider Credentials By TypeBRead-onlyIdempotent
List all credentials for a given provider type.
Inputs: provider_type (required).
lyzr_list_provider_credentials_by_userList Provider Credentials By User And TypeARead-onlyIdempotent
List a user's credentials for a given provider type and provider id.
Inputs: provider_id (required), provider_type (required).
lyzr_list_resource_audit_logsList Resource Audit LogsBRead-onlyIdempotent
Get audit logs for a specific resource type and ID, with optional filters.
Inputs: resource_id (required), resource_type (required), limit, action, +4 more.
lyzr_list_session_audit_logsList Session Audit LogsARead-onlyIdempotent
Get audit logs for a specific session ID.
Inputs: session_id (required), limit, offset.
lyzr_list_toolsList ToolsBRead-onlyIdempotent
List the caller's registered tools.
Inputs: no named parameters.
lyzr_list_tracesList TracesARead-onlyIdempotent
List agent execution traces, optionally filtered by time range, agent, session, user, or customer.
Inputs: limit, offset, agent_id, end_time, +5 more.
lyzr_list_user_asset_filtersList User Asset Filter FacetsARead-onlyIdempotent
List the available filter facets (providers, models, owners, tags, etc.) for user assets.
Inputs: no named parameters.
lyzr_list_user_assetsList User AssetsARead-onlyIdempotent
List the caller's user assets (agents, folders, workflows, etc.) with filtering, sorting, and pagination.
Inputs: page, tags, type, limit, +12 more.
lyzr_list_user_audit_logsList User Audit LogsARead-onlyIdempotent
Get audit logs for a specific user by ID, with optional filters.
Inputs: user_id (required), limit, action, offset, +4 more.
lyzr_list_voicesList VoicesARead-onlyIdempotent
List the available ElevenLabs voices configuration.
Inputs: no named parameters.
lyzr_list_workflowsList WorkflowsARead-onlyIdempotent
List all workflows.
Inputs: no named parameters.
lyzr_livesource_addAdd Live SourceA
Add a live source to a knowledge base. Returns the created live source.
Inputs: name (required), rag_id (required), source_type (required), metadata, +4 more.
lyzr_livesource_ext_browse_childrenBrowse SharePoint Drive ChildrenARead-onlyIdempotent
List files/folders within a SharePoint drive folder.
Inputs: site_url (required), drive_name (required), credential_id (required), folder_path.
lyzr_livesource_ext_browse_drivesBrowse SharePoint DrivesARead-onlyIdempotent
List drives within a SharePoint site.
Inputs: site_url (required), credential_id (required).
lyzr_livesource_ext_browse_sitesBrowse SharePoint SitesARead-onlyIdempotent
List SharePoint sites accessible via a stored credential.
Inputs: credential_id (required).
lyzr_livesource_ext_sync_permissionsSync Live Source PermissionsAIdempotent
Sync access permissions for a live source (e.g. re-check SharePoint ACLs).
Inputs: rag_id (required), live_source_id (required).
lyzr_livesource_ext_validate_accessValidate SharePoint Site AccessARead-onlyIdempotent
Validate that a credential has access to the given SharePoint site/drive URLs.
Inputs: site_urls (required), credential_id (required).
lyzr_livesource_ext_webhook_getLive Source Webhook ValidationBRead-onlyIdempotent
Webhook subscription validation handshake for live source notifications.
Inputs: no named parameters.
lyzr_livesource_ext_webhook_postLive Source Webhook NotificationC
Send a live source webhook notification payload (e.g. from a SharePoint change feed).
Inputs: payload (required).
lyzr_livesource_getGet Live SourceBRead-onlyIdempotent
Fetch a live source by id.
Inputs: rag_id (required), live_source_id (required).
lyzr_livesource_listList Live SourcesARead-onlyIdempotent
List live sources for a knowledge base.
Inputs: rag_id (required).
lyzr_livesource_list_credentialsList Live Source CredentialsARead-onlyIdempotent
List credentials available for use with live sources.
Inputs: no named parameters.
lyzr_livesource_pausePause Live SourceAIdempotent
Pause syncing for a live source.
Inputs: rag_id (required), live_source_id (required).
lyzr_livesource_removeRemove Live SourceADestructiveIdempotent
Permanently remove a live source from a knowledge base.
Inputs: rag_id (required), live_source_id (required).
lyzr_livesource_repointRepoint Live SourceA
Repoint a live source to a new credential. Existing webhook subscriptions are re-created.
Inputs: rag_id (required), live_source_id (required), new_credential_id (required).
lyzr_livesource_resumeResume Live SourceAIdempotent
Resume syncing for a paused live source.
Inputs: rag_id (required), live_source_id (required).
lyzr_livesource_syncSync Live SourceB
Trigger a sync for a live source.
Inputs: rag_id (required), live_source_id (required).
lyzr_log_auth_eventLog Auth EventA
Log a login or logout event from the frontend into the audit log.
Inputs: event_type (required), metadata, user_email.
lyzr_memory_addAdd MemoryA
Store conversation messages in Cognis memory. At least one of owner_id/agent_id/session_id is required.
Inputs: messages (required), agent_id, owner_id, session_id.
lyzr_memory_deleteDelete MemoryBDestructiveIdempotent
Delete a memory record by id.
Inputs: memory_id (required), owner_id.
lyzr_memory_getGet MemoryARead-onlyIdempotent
Fetch a single memory record by id.
Inputs: memory_id (required), owner_id.
lyzr_memory_listList MemoriesBRead-onlyIdempotent
List stored memories for the given identifiers.
Inputs: limit, offset, agent_id, owner_id, +2 more.
lyzr_memory_searchSearch MemoryBRead-onlyIdempotent
Semantic search over stored memories.
Inputs: query (required), limit, agent_id, owner_id, +2 more.
lyzr_memory_updateUpdate MemoryBIdempotent
Update a memory record's content and/or metadata.
Inputs: memory_id (required), content, owner_id.
lyzr_memprovider_aws_agentcore_statusGet AWS AgentCore Provisioning StatusARead-onlyIdempotent
Get the provisioning status of an AWS AgentCore memory resource.
Inputs: credential_id (required).
lyzr_memprovider_delete_aws_agentcore_resourceDelete AWS AgentCore Memory ResourceADestructiveIdempotent
Permanently delete the AWS AgentCore memory resource for a credential.
Inputs: credential_id (required).
lyzr_memprovider_get_providerGet Memory ProviderARead-onlyIdempotent
Fetch a single memory provider by id.
Inputs: provider_id (required).
lyzr_memprovider_list_aws_agentcore_resourcesList AWS AgentCore Memory ResourcesBRead-onlyIdempotent
List the AWS AgentCore memory resources for a credential.
Inputs: credential_id (required).
lyzr_memprovider_list_providersList Memory ProvidersARead-onlyIdempotent
List the memory providers available for agents.
Inputs: no named parameters.
lyzr_memprovider_mem0_statusGet Mem0 StatusBRead-onlyIdempotent
Get the status of a Mem0 memory provider credential.
Inputs: credential_id (required).
lyzr_memprovider_provision_aws_agentcore_memoryProvision AWS AgentCore MemoryB
Provision a new AWS AgentCore memory resource.
Inputs: memory_name (required), credential_id (required), memory_strategy, event_expiry_days.
lyzr_memprovider_supermemory_statusGet Supermemory StatusBRead-onlyIdempotent
Get the status of a Supermemory memory provider credential.
Inputs: credential_id (required).
lyzr_memprovider_use_existing_aws_agentcore_memoryUse Existing AWS AgentCore MemoryB
Attach an existing AWS AgentCore memory resource to a credential.
Inputs: memory_id (required), credential_id (required).
lyzr_memprovider_validate_aws_agentcoreValidate AWS AgentCore CredentialsARead-onlyIdempotent
Validate AWS AgentCore credentials and list associated memories.
Inputs: credential_id (required).
lyzr_memprovider_validate_mem0Validate Mem0 CredentialsARead-onlyIdempotent
Validate Mem0 memory provider credentials.
Inputs: credential_id (required).
lyzr_memprovider_validate_supermemoryValidate Supermemory CredentialsARead-onlyIdempotent
Validate Supermemory memory provider credentials.
Inputs: credential_id (required).
lyzr_ops_export_report_csvExport Ops Report CSVB
Export an operations report as CSV for a date range.
Inputs: end_date (required), start_date (required).
lyzr_ops_generate_reportGenerate Ops ReportC
Generate an operations report for a date range.
Inputs: end_date (required), start_date (required).
lyzr_ops_get_agent_tool_logsGet Agent Tool LogsARead-onlyIdempotent
Fetch agent tool logs for a trace/run/log/feature combination.
Inputs: log_id (required), run_id (required), feature (required), trace_id (required).
lyzr_ops_get_dashboardGet Ops DashboardARead-onlyIdempotent
Fetch the operations dashboard for a date range, optionally scoped to one agent.
Inputs: end_date (required), start_date (required), agent_id.
lyzr_ops_get_grouped_logsGet Grouped Activity LogsARead-onlyIdempotent
Fetch grouped activity logs for a trace/run/log combination.
Inputs: log_id (required), run_id (required), trace_id (required), page, +1 more.
lyzr_ops_get_traceGet Ops TraceBRead-onlyIdempotent
Fetch a single execution trace by id.
Inputs: trace_id (required).
lyzr_ops_get_trace_runGet Ops Trace RunBRead-onlyIdempotent
Fetch a specific run within a trace.
Inputs: run_id (required), trace_id (required).
lyzr_ops_get_tracesList Ops TracesARead-onlyIdempotent
List execution traces, optionally filtered by agent and date range.
Inputs: page, count, limit, agent_id, +2 more.
lyzr_parse_csvParse CSV FileB
Parse a CSV file into text chunks.
Inputs: filename (required), source_column (required), file_content_base64 (required), mime_type, +1 more.
lyzr_parse_docxParse DOCX FileA
Parse a DOCX (Word) file into text chunks.
Inputs: filename (required), file_content_base64 (required), s3_key, mime_type, +6 more.
lyzr_parse_imageParse Image FileA
Parse an image file (e.g. via OCR/VLM) into text chunks.
Inputs: filename (required), file_content_base64 (required), mime_type, chunk_size, +3 more.
lyzr_parse_pdfParse PDF FileB
Parse a PDF file into text chunks.
Inputs: filename (required), file_content_base64 (required), s3_key, mime_type, +6 more.
lyzr_parse_pptxParse PPTX FileA
Parse a PPTX (PowerPoint) file into text chunks.
Inputs: filename (required), file_content_base64 (required), mime_type, chunk_size, +3 more.
lyzr_parse_textParse TextB
Parse raw text into chunks (no knowledge base required).
Inputs: text (required), chunk_size, extra_info, extra_fields, +2 more.
lyzr_parse_txtParse TXT FileC
Parse a plain-text file into text chunks.
Inputs: filename (required), file_content_base64 (required), s3_key, mime_type, +6 more.
lyzr_parse_websiteParse WebsiteA
Parse a website URL into chunks (no knowledge base required).
Inputs: url (required), chunk_size, extra_info, extra_fields, +2 more.
lyzr_parse_website_apifyParse Website (Apify)A
Parse a website URL into chunks using the Apify crawler backend.
Inputs: url (required), chunk_size, extra_info, extra_fields, +2 more.
lyzr_parse_xlsxParse XLSX FileB
Parse an XLSX (Excel) file into text chunks.
Inputs: filename (required), file_content_base64 (required), mime_type, chunk_size, +3 more.
lyzr_publish_agentsPublish AgentsC
Publish one or more agents with a given access level.
Inputs: agent_ids (required), access_level.
lyzr_rag_get_doc_contentGet RAG Document ContentARead-onlyIdempotent
Get the full stored content (chunks) of a single document source in a knowledge base.
Inputs: rag_id (required), source (required), limit.
lyzr_rag_source_auth_aci_handoffRAG Source Auth ACI HandoffBRead-only
Complete the ACI source-auth hand-off callback using the returned state token.
Inputs: state (required).
lyzr_rag_source_auth_statusRAG Source Auth StatusBRead-onlyIdempotent
Get the connection/auth status of external data sources configured for a knowledge base.
Inputs: rag_id (required).
lyzr_rai_create_policyCreate RAI PolicyA
Create a Responsible-AI guardrail policy (toxicity, prompt-injection, NSFW, banned topics).
Inputs: name (required), nsfw_check, description, banned_topics, +3 more.
lyzr_rai_delete_policyDelete RAI PolicyBDestructiveIdempotent
Delete a guardrail policy by id.
Inputs: policy_id (required).
lyzr_rai_get_policyGet RAI PolicyARead-onlyIdempotent
Fetch a guardrail policy by id.
Inputs: policy_id (required).
lyzr_rai_list_policiesList RAI PoliciesARead-onlyIdempotent
List your Responsible-AI guardrail policies.
Inputs: no named parameters.
lyzr_reassign_agentReassign AgentA
Reassign ownership of an agent to another user by email.
Inputs: agent_id (required), target_email (required).
lyzr_refresh_credit_cacheRefresh Credit CacheA
Force a refresh of the server-side credit-cost cache.
Inputs: x_server_token.
lyzr_refresh_tool_credential_statusRefresh Tool Credential StatusBIdempotent
Refresh/recheck the status of a tool credential.
Inputs: credential_id (required).
lyzr_remove_channel_agent_routeRemove Channel Agent RouteADestructiveIdempotent
Remove an agent route from a channel.
Inputs: agent_id (required), channel_id (required).
lyzr_report_get_statusGet Report StatusARead-onlyIdempotent
Get the status of a report job, including a download_url once it is ready.
Inputs: job_id (required).
lyzr_report_listList ReportsARead-onlyIdempotent
List report jobs, optionally filtered by type or status.
Inputs: skip, limit, status, report_type.
lyzr_report_usage_by_agentRequest Usage-By-Agent ReportA
Request an async usage report broken down by agent. Returns a job_id; poll with lyzr_report_get_status.
Inputs: timeframe (required), model, end_date, group_by, +4 more.
lyzr_report_usage_by_modelRequest Usage-By-Model ReportA
Request an async usage report broken down by model. Returns a job_id; poll with lyzr_report_get_status.
Inputs: timeframe (required), model, end_date, group_by, +3 more.
lyzr_report_usage_by_sub_accountRequest Usage-By-Sub-Account ReportA
Request an async usage report broken down by sub-account. Returns a job_id; poll with lyzr_report_get_status.
Inputs: timeframe (required), end_date, group_by, start_date, +1 more.
lyzr_report_usage_by_userRequest Usage-By-User ReportA
Request an async usage report broken down by user. Returns a job_id; poll with lyzr_report_get_status.
Inputs: timeframe (required), end_date, group_by, start_date, +1 more.
lyzr_resolve_asset_by_sourceResolve Asset By SourceBRead-onlyIdempotent
Resolve an asset from a knowledge base document's source path.
Inputs: rag_id (required), source (required).
lyzr_resolve_llm_credentialResolve LLM CredentialBRead-onlyIdempotent
Resolve an LLM credential by credential or provider id.
Inputs: provider_id, credential_id.
lyzr_run_usage_alertsRun Usage Alert PassB
Trigger a usage alert evaluation pass on the server. Optionally pass a server token header.
Inputs: x_server_token.
lyzr_schedule_createCreate ScheduleC
Schedule an agent to run on a cron expression (5 fields: minute hour day month weekday).
Inputs: user_id (required), agent_id (required), cron_expression (required), message, +3 more.
lyzr_schedule_deleteDelete ScheduleBDestructiveIdempotent
Delete a schedule by id.
Inputs: schedule_id (required).
lyzr_schedule_getGet ScheduleARead-onlyIdempotent
Fetch a schedule by id.
Inputs: schedule_id (required).
lyzr_schedule_listList SchedulesARead-onlyIdempotent
List schedules (optionally filtered by user_id).
Inputs: user_id.
lyzr_schedule_pausePause ScheduleBIdempotent
Pause a schedule.
Inputs: schedule_id (required).
lyzr_schedule_resumeResume ScheduleCIdempotent
Resume a paused schedule.
Inputs: schedule_id (required).
lyzr_schedule_triggerTrigger Schedule NowB
Run a schedule immediately (out of band).
Inputs: schedule_id (required).
lyzr_search_user_assetsSearch User AssetsARead-onlyIdempotent
Search the caller's user assets by query string, with the same filtering, sorting, and pagination options as listing.
Inputs: q (required), page, tags, type, +13 more.
lyzr_semantic_model_connect_databaseConnect Database To RAG ConfigA
Connect a database to a RAG config for semantic-model table documentation.
Inputs: database_id (required), rag_config_id (required).
lyzr_semantic_model_create_documentation_agentCreate Documentation AgentA
Create a new semantic-model documentation agent used to generate table/column descriptions.
Inputs: name (required), top_p (required), model_id (required), provider_id (required), +2 more.
lyzr_semantic_model_get_descriptionsGet Table DescriptionsARead-onlyIdempotent
Get saved table and column descriptions for a table in a database.
Inputs: table_name (required), database_id (required), rag_config_id (required).
lyzr_semantic_model_get_task_statusGet Semantic Model Task StatusARead-onlyIdempotent
Get the status of a semantic-model background task (e.g. save_documentation_task).
Inputs: task_id (required).
lyzr_semantic_model_list_documentation_agentsList Documentation AgentsARead-onlyIdempotent
List semantic-model documentation agents.
Inputs: no named parameters.
lyzr_semantic_model_list_tablesList TablesARead-onlyIdempotent
List table names available in a connected database for a RAG config.
Inputs: database_id (required), rag_config_id (required).
lyzr_semantic_model_remove_documentationRemove Table DocumentationBDestructiveIdempotent
Remove saved documentation for a table.
Inputs: table_name (required), rag_config_id (required).
lyzr_semantic_model_remove_documentation_taskRemove Documentation TaskADestructiveIdempotent
Remove a background documentation-save task for a table.
Inputs: table_name (required), rag_config_id (required).
lyzr_semantic_model_save_documentationSave Table DocumentationA
Save table/column descriptions and a table preview for a table (synchronous).
Inputs: table_name (required), descriptions (required), rag_config_id (required), table_preview (required).
lyzr_semantic_model_save_documentation_taskSave Table Documentation (Background Task)A
Save table/column descriptions and a table preview for a table as a background task. Returns a task_id to poll with lyzr_semantic_model_get_task_status.
Inputs: table_name (required), descriptions (required), rag_config_id (required), table_preview (required).
lyzr_semantic_model_table_previewPreview TableARead-onlyIdempotent
Preview rows of a table in a connected database.
Inputs: table_name (required), database_id (required), rag_config_id (required).
lyzr_send_a2a_jsonrpcSend A2A JSON-RPC RequestA
Send a raw JSON-RPC 2.0 request to a served A2A agent's endpoint.
Inputs: request (required), agent_id (required).
lyzr_session3_ancestryGet Session Ancestry (v3)ARead-onlyIdempotent
Get the ancestry chain of a v3 session, root to leaf.
Inputs: session_id (required).
lyzr_session3_branchBranch Session (v3)A
Branch a v3 session off from a specific message.
Inputs: session_id (required), from_message_id (required), metadata, branch_name, +1 more.
lyzr_session3_createCreate Session (v3)B
Create a new v3 session, optionally tied to an agent.
Inputs: session_id (required), source, agent_id, metadata.
lyzr_session3_deleteDelete Session (v3)ADestructiveIdempotent
Permanently delete a v3 session by id.
Inputs: session_id (required).
lyzr_session3_getGet Session (v3)ARead-onlyIdempotent
Fetch a v3 session by id.
Inputs: session_id (required).
lyzr_session3_listList Sessions (v3)ARead-onlyIdempotent
List v3 sessions, optionally filtered by agent or source.
Inputs: limit, offset, source, agent_id.
lyzr_session3_list_branchesList Session Branches (v3)ARead-onlyIdempotent
List the branches created from a v3 session.
Inputs: session_id (required).
lyzr_session3_list_messagesList Session Messages (v3)ARead-onlyIdempotent
List messages in a v3 session, with optional filters.
Inputs: session_id (required), role, after, limit, +1 more.
lyzr_session3_treeGet Session Tree (v3)ARead-onlyIdempotent
Get the full branch tree rooted at a v3 session.
Inputs: session_id (required).
lyzr_session3_updateUpdate Session (v3)A
Update a v3 session's metadata.
Inputs: metadata (required), session_id (required).
lyzr_session_conversationGet Session ConversationBRead-onlyIdempotent
Fetch a session's conversation.
Inputs: session_id (required).
lyzr_session_createCreate SessionC
Create a new session.
Inputs: user_id (required), agent_id, metadata, extra_fields.
lyzr_session_create_for_agentCreate Session for AgentA
Create a new session bound to a specific agent.
Inputs: user_id (required), agent_id (required), metadata, extra_fields.
lyzr_session_deleteDelete SessionBDestructiveIdempotent
Permanently delete a session by id.
Inputs: session_id (required).
lyzr_session_getGet SessionBRead-onlyIdempotent
Fetch a session by id.
Inputs: session_id (required), timeout.
lyzr_session_historyGet Session HistoryBRead-onlyIdempotent
Fetch a session's message history.
Inputs: session_id (required), unix.
lyzr_session_history_by_agentGet Session History by AgentBRead-onlyIdempotent
Fetch a session's message history for a specific agent.
Inputs: agent_id (required), session_id (required), unix.
lyzr_session_summaryGet Session SummaryBRead-onlyIdempotent
Fetch a session's summary.
Inputs: session_id (required).
lyzr_session_updateUpdate SessionCIdempotent
Update an existing session by id.
Inputs: user_id (required), session_id (required), agent_id, metadata, +1 more.
lyzr_set_agent_lockSet Agent LockAIdempotent
Lock or unlock an agent, optionally scoped to an environment.
Inputs: agent_id (required), is_locked (required), environment.
lyzr_set_agent_statusSet Agent StatusBIdempotent
Activate or deactivate an agent.
Inputs: agent_id (required), is_active (required).
lyzr_sharing_check_accessCheck Shared AccessARead-onlyIdempotent
Check whether a user (via their org) has the required access level to a resource.
Inputs: org_id (required), user_id (required), resource_id (required), resource_type (required), +1 more.
lyzr_sharing_create_groupCreate Sharing GroupA
Create a sharing group rooted at a resource, controlling who can access it.
Inputs: root_resource_id (required), root_resource_type (required), agent_ids, shared_with, +3 more.
lyzr_sharing_delete_groupDelete Sharing GroupADestructiveIdempotent
Permanently delete a sharing group by id.
Inputs: group_id (required).
lyzr_sharing_get_groupGet Sharing GroupBRead-onlyIdempotent
Fetch a sharing group by id.
Inputs: group_id (required).
lyzr_sharing_get_resource_groupsGet Resource's Sharing GroupsBRead-onlyIdempotent
List the sharing groups a given resource belongs to.
Inputs: resource_id (required), resource_type (required), root_tree.
lyzr_sharing_list_accessibleList Accessible ResourcesARead-onlyIdempotent
List resources of a given type that a user (via their org) has shared access to.
Inputs: org_id (required), user_id (required), resource_type (required).
lyzr_sharing_list_groupsList Sharing GroupsARead-onlyIdempotent
List sharing groups, paginated.
Inputs: page, limit.
lyzr_sharing_reconcile_indexesReconcile Sharing IndexesA
Trigger a reconciliation pass over all sharing indexes.
Inputs: no named parameters.
lyzr_sharing_refresh_groupRefresh Sharing GroupA
Refresh a sharing group's cached shared-resource index.
Inputs: group_id (required).
lyzr_sharing_update_groupUpdate Sharing GroupCIdempotent
Update a sharing group's access levels or shared users.
Inputs: group_id (required), shared_with, access_level, org_access_level.
lyzr_simple_chat_completionsSimple Chat CompletionsA
OpenAI-compatible chat/completions endpoint that selects the model/provider directly, without a pre-created agent.
Inputs: model (required), messages (required), top_p, session_id, +3 more.
lyzr_start_sessionStart SessionBIdempotent
Start (or resume) an inference session.
Inputs: session_id (required).
lyzr_start_taskStart Lyzr TaskA
Submit a long-running task to a Lyzr agent. Returns a task_id to poll with lyzr_get_task_status.
Inputs: message (required), agent_id (required), user_id, session_id.
lyzr_stop_sessionStop SessionBIdempotent
Stop an active inference session.
Inputs: session_id (required).
lyzr_stream_chatStream Chat with Lyzr AgentA
Send a message to a Lyzr agent over the streaming endpoint; streams progress and returns the full response.
Inputs: message (required), agent_id (required), user_id, session_id.
lyzr_submit_chat_taskSubmit Chat TaskA
Submit a chat message to an agent as an async task; returns a task_id to poll.
Inputs: agent_id (required), session_id (required), message, user_id, +1 more.
lyzr_test_tool_credentialTest Tool CredentialB
Run a live test of a tool credential.
Inputs: credential_id (required).
lyzr_trigger_workflow_with_fileTrigger Workflow With FileB
Trigger a workflow run by uploading a file (multipart/form-data).
Inputs: flow_id (required), file_name (required), file_content (required), is_base64, +1 more.
lyzr_update_a2a_agentUpdate A2A AgentBIdempotent
Update an existing A2A agent record.
Inputs: agent_id (required), base_url (required), name, skills, +12 more.
lyzr_update_agentUpdate Lyzr AgentAIdempotent
Update fields (name/role/goal/instructions/temperature/description) on an existing Lyzr agent.
Inputs: agent_id (required), goal, name, role, +3 more.
lyzr_update_artifactUpdate ArtifactAIdempotent
Update fields on an existing artifact, scoped to a user and session.
Inputs: user_id (required), session_id (required), artifact_id (required), data, +4 more.
lyzr_update_asset_parsing_statusUpdate Asset Parsing StatusBIdempotent
Update the parsing status/metadata of an asset by id.
Inputs: update (required), asset_id (required).
lyzr_update_contextUpdate ContextAIdempotent
Update a context's name and/or value by id.
Inputs: context_id (required), name, value.
lyzr_update_feature_flagUpdate Feature FlagC
Update an existing feature flag's fields.
Inputs: key (required), is_visible, description, is_accessible, +3 more.
lyzr_update_file_upload_credentialUpdate File-Upload Provider CredentialCIdempotent
Update a file-upload provider credential, optionally replacing its files.
Inputs: update_data (required), credential_id (required), files.
lyzr_update_moduleUpdate ModuleB
Update an existing sidebar module's fields.
Inputs: key (required), beta, icon, name, +20 more.
lyzr_update_org_llm_fallbacksUpdate Org LLM FallbacksAIdempotent
Replace the organization's LLM fallback chain with the given ordered list of fallback entries.
Inputs: fallbacks (required).
lyzr_update_providerUpdate ProviderBIdempotent
Update an existing provider's form/metadata.
Inputs: form (required), type (required), meta_data (required), provider_id (required).
lyzr_update_provider_credentialUpdate Provider CredentialBIdempotent
Update an existing provider credential by id.
Inputs: name (required), type (required), user_id (required), meta_data (required), +3 more.
lyzr_update_toolUpdate ToolBIdempotent
Update fields on an existing tool by id.
Inputs: update (required), tool_id (required).
lyzr_update_workflowUpdate WorkflowBIdempotent
Update fields on an existing workflow.
Inputs: flow_id (required), api_key, flow_data, flow_name.
lyzr_upload_assetUpload AssetA
Upload one or more files as assets, optionally parsing/chunking them for RAG. Files are base64-encoded in the request.
Inputs: files (required), end_page, vlm_model, enable_vlm, +8 more.
lyzr_widget_streamWidget Stream ChatA
Send a message to the pre-configured embeddable widget agent and stream back the response. No agent_id or credentials are accepted — the target agent is fixed server-side.
Inputs: message (required), session_id (required).
lyzr_world_model_add_personasAdd World Model PersonasB
Add one or more personas to a world model.
Inputs: personas (required), world_model_id (required).
lyzr_world_model_add_scenariosAdd World Model ScenariosB
Add one or more scenarios to a world model.
Inputs: scenarios (required), world_model_id (required).
lyzr_world_model_add_test_casesAdd World Model Test CasesA
Add one or more test cases to a world model.
Inputs: test_cases (required), world_model_id (required).
lyzr_world_model_createCreate World ModelA
Create a world model (a simulation clone) from a source agent. Returns the world_model_id.
Inputs: source_agent_id (required), name.
lyzr_world_model_create_evaluation_runCreate World Model Evaluation RunB
Create a new evaluation run for an agent against a World Model's test cases.
Inputs: agent_id (required), run_name (required), agent_name (required), world_model_id (required), +10 more.
lyzr_world_model_dashboard_by_world_modelGet World Model DashboardARead-onlyIdempotent
Fetch the dashboard summary for a single World Model.
Inputs: world_model_id (required).
lyzr_world_model_dashboard_overviewGet World Model Dashboard OverviewARead-onlyIdempotent
Fetch the overall World Model dashboard summary across all world models.
Inputs: no named parameters.
lyzr_world_model_deleteDelete World ModelADestructiveIdempotent
Permanently delete a world model by id.
Inputs: world_model_id (required).
lyzr_world_model_delete_personaDelete World Model PersonaBDestructiveIdempotent
Delete a persona from a world model.
Inputs: persona_id (required), world_model_id (required).
lyzr_world_model_delete_scenarioDelete World Model ScenarioBDestructiveIdempotent
Delete a scenario from a world model.
Inputs: scenario_id (required), world_model_id (required).
lyzr_world_model_delete_test_caseDelete World Model Test CaseADestructiveIdempotent
Delete a test case from a world model.
Inputs: test_case_id (required), world_model_id (required).
lyzr_world_model_getGet World ModelARead-onlyIdempotent
Fetch a world model by id.
Inputs: world_model_id (required).
lyzr_world_model_get_evaluation_runGet World Model Evaluation RunARead-onlyIdempotent
Fetch a single evaluation run by its id.
Inputs: run_id (required).
lyzr_world_model_list_by_agentList World Models By AgentARead-onlyIdempotent
List all world models built from a given source agent.
Inputs: agent_id (required).
lyzr_world_model_list_evaluation_runsList World Model Evaluation RunsARead-onlyIdempotent
List all evaluation runs created for a given World Model.
Inputs: world_model_id (required).
lyzr_world_model_list_personasList World Model PersonasBRead-onlyIdempotent
List the personas defined on a world model.
Inputs: world_model_id (required).
lyzr_world_model_list_scenariosList World Model ScenariosBRead-onlyIdempotent
List the scenarios defined on a world model.
Inputs: world_model_id (required).
lyzr_world_model_list_test_casesList World Model Test CasesARead-onlyIdempotent
List the test cases defined on a world model.
Inputs: world_model_id (required).
lyzr_world_model_update_personaUpdate World Model PersonaBIdempotent
Update a persona on a world model.
Inputs: name (required), persona_id (required), description (required), world_model_id (required).
lyzr_world_model_update_scenarioUpdate World Model ScenarioAIdempotent
Update a scenario on a world model.
Inputs: name (required), description (required), scenario_id (required), world_model_id (required).
lyzr_world_model_update_test_caseUpdate World Model Test CaseBIdempotent
Update a test case on a world model.
Inputs: name (required), user_input (required), test_case_id (required), world_model_id (required), +3 more.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 423 tools
There are numerous near-identical clusters (e.g., lyzr_kg_train_text vs lyzr_kg_ext_train_text_neo4j, session_create/session_create_for_agent/session3_create, and multiple session history variants) whose boundaries are unclear from names alone. The 423-tool surface forces agents to choose between duplicates, v3/ext/task/internal variants, and deprecated endpoints with overlapping purposes.
All tools share the lyzr_ prefix and snake_case, but the verb/noun ordering is inconsistent: lyzr_create_agent, lyzr_kb_create, lyzr_session_create, lyzr_agent_create_single_task, and lyzr_report_usage_by_model all use different constructions. There are also bare nouns like lyzr_agent_sessions and single verbs like lyzr_extract, so no reliable naming pattern predicts a tool's role.
423 tools is an extreme count for an MCP server, far beyond the 3-15 well-scoped range and even beyond the 25+ threshold. Even if the underlying platform is broad, exposing this much surface destroys navigability and makes correct tool selection impractical.
The tool surface provides thorough CRUD and lifecycle coverage for agents, sessions, knowledge bases, credentials, workflows, git repos, world models, A2A agents, sharing, reports, and more, including async task variants and status polling. There are no obvious dead ends; most resources have create/read/update/delete plus operational triggers.
Related MCP Connectors
An MCP server that provides an API to LLMs to manage their JumpCloud resources.
MCP server for building and testing AI agents with multi-model experimentation and insights.
MCP server for AI agents to plan, verify, and deploy Cloudflare-native apps.
MCP server for your apps' tools and custom tools, plus hosted AI agents and approval-gated workflows
Related MCP Servers
- FlicenseNot gradedqualityDmaintenanceA production-ready MCP server built with FastAPI, providing an enhanced tool registry for creating, managing, and documenting AI tools for Large Language Models (LLMs).34-
- AlicenseAqualityBmaintenanceMCP server that exposes any LynxPrompt instance to LLMs, enabling browsing, searching, and managing AI configuration blueprints and prompt hierarchies.10358 npm3GPL 3.0
- AlicenseAqualityDmaintenanceMCP server that exposes 300+ AI agents as tools via a single API key. Supports listing agents, invoking any agent with chat-completion style messages, checking agent health, and retrieving platform statistics.54MIT

Agorus MCP Serverofficial
AlicenseNot gradedqualityDmaintenanceMCP server for the Agorus AI agent marketplace, exposing API operations as tools for LLMs to discover, contract, and interact with agents and services.13 npmMIT