casemgr-mcp
casemgr-mcp
A shared workspace your AI agent actually writes to — 184 MCP tools.
It creates notes when it finds things, plans tasks when work appears, and marks them done. You review and edit. Persistent memory + semantic search across a graph of notes, tasks, and files.
This package is the stdio-to-HTTP bridge for the CaseMgr MCP server — use it with any MCP client that supports stdio transport (Claude Desktop, Cursor, etc.). For Claude Code, see below — it supports HTTP natively and does not need this bridge.
Quick Start
npm install -g casemgr-mcpGet an API token at casemgr.systems/tokens.
Related MCP server: Elasticflow MCP Server
MCP Client Config
Claude Desktop / Cursor
Add to your MCP settings:
{
"mcpServers": {
"casemgr": {
"command": "casemgr-mcp",
"env": {
"CASEMGR_API_TOKEN": "your_token_here"
}
}
}
}Claude Code (CLI)
Claude Code authenticates via OAuth automatically — no token needed. Just add the server URL:
{
"mcpServers": {
"casemgr": {
"url": "https://casemgr.systems/mcp"
}
}
}If you prefer token auth, add "headers": { "Authorization": "Bearer your_token_here" }.
Smithery
Install directly via Smithery — one-click setup for Claude Code, Cursor, Codex, and more.
What is CaseMgr?
A shared, persistent workspace for you and your AI agent — 184 MCP tools across a graph of notes, tasks, files, calendar, and agent presence.
Your agent creates notes when it finds things. Plans tasks when there's work to do. Marks them done as it finishes. You review, edit, and add your own. Nothing is ephemeral — every session resumes exactly where the last one left off, and semantic search spans everything either of you ever wrote.
What's in the workspace:
Persistent memory — every session resumes where the last one left off; nothing is ephemeral
Bidirectional graph — agent and user both read and write notes, tasks, bookmarks, files, todos
Semantic search — natural-language search across everything either of you ever wrote
Multi-agent task queue — dispatch, claim, and complete work items across multiple agents
Event-driven workflows — CMMN stages with sentry-gated execution and auto-lifecycle management
Reusable templates — case plan models with publish, version, and instantiate
File management — upload, download, versioning, content-aware embeddings
Workspaces & worktrees — organize cases, link git worktrees across machines
Time tracking & billing — durations, expenses, invoices with LaTeX/PDF generation
Environment Variables
Variable | Required | Default | Description |
| Yes (stdio) | — | API token from casemgr.systems/tokens. Required for stdio clients (Claude Desktop, Cursor). Not needed for Claude Code, which uses OAuth. |
| No |
| MCP endpoint URL (for self-hosted instances) |
License
MIT
Available Tools
205 toolsagents-deregisteragents-deregisterA
Marks an agent session offline. The vertex is preserved so the next register call can refresh it; the live presence record is removed.
Inputs: agent_session_id (required).
agents-getagents-getB
Fetches a single agent session by RID, with live_present indicating live MCP connection state.
Inputs: agent_session_id (required).
agents-heartbeatagents-heartbeatA
Refreshes an agent session's heartbeat. Bumps last_heartbeat_at to keep the agent marked online. Call periodically (~30s) to avoid being swept offline.
Inputs: agent_session_id (required), current_case_id.
agents-listagents-listA
Lists agent sessions owned by the current user. Each entry includes a live_present flag indicating whether the agent has a live ETS frame (i.e., currently connected via MCP).
Inputs: machine_id, status.
agents-registeragents-registerA
Registers an MCP agent session. Upserts on (user_id, machine_id, alias) — reconnects update the existing row. Returns the agent_session_id used by other tools.
Inputs: alias (required), machine_id (required), capabilities, current_case_id, +2 more.
ai-cancel_work_itemai-cancel_work_itemA
Cancels a pending work item (pending → cancelled). Cannot cancel items that are already processing.
Inputs: work_item_id (required).
ai-claim_and_complete_work_itemai-claim_and_complete_work_itemA
Atomically claims a pending work item and completes it. Use this when the agent already knows the result and doesn't need a separate processing step — collapses the (claim → complete) two-call sequence into one. Idempotent: if the item is already in 'processing', skips the claim and just completes. Items in terminal states (completed/failed/cancelled) return an error.
Inputs: work_item_id (required), output_content, output_name, result.
ai-claim_work_itemai-claim_work_itemA
Claims a pending work item for processing (pending → processing). Call this before starting to process a work item.
Inputs: work_item_id (required).
ai-complete_work_itemai-complete_work_itemA
Marks a work item as completed with optional result and output. If output_item_type was configured, creates the output item. Triggers connected sentry if configured. Item must be in 'processing' status — call ai-claim_work_item first, or use ai-claim_and_complete_work_item to do both atomically.
Inputs: work_item_id (required), output_content, output_name, result.
ai-count_pendingai-count_pendingA
Gets the count of pending AI work items. Useful for checking if there's work available.
Inputs: agent_type.
ai-fail_work_itemai-fail_work_itemB
Marks a work item as failed with an error message.
Inputs: error (required), work_item_id (required).
ai-get_work_itemai-get_work_itemA
Gets a specific AI work item with full details including prompt, input_items, and configuration.
Inputs: work_item_id (required).
ai-list_work_itemsai-list_work_itemsA
Lists AI work items. By default returns pending items sorted by priority. Use this to find work for AI agents to process.
Inputs: agent_type, case_id, limit, status.
billing-generate_invoice_pdfbilling-generate_invoice_pdfA
Generates a PDF invoice and stores it as a file item in the specified case.
Inputs: case_id (required), invoice_id (required).
billing-generate_invoice_texbilling-generate_invoice_texB
Generates a LaTeX invoice document and stores it as a file item in the specified case.
Inputs: case_id (required), invoice_id (required).
calendar-get_or_createcalendar-get_or_createC
Gets or creates the main calendar
Inputs: name.
calendar-get_or_create_daycalendar-get_or_create_dayA
Gets or creates a calendar day (also creates month if needed)
Inputs: date (required).
cases-add_documentationcases-add_documentationA
Adds a documentation entry to a case following the CMMN tCmmnElement pattern. Returns the updated case with all documentation entries.
Inputs: case_id (required), text (required), text_format.
cases-add_tagcases-add_tagB
Adds a tag to a case for categorization
Inputs: case_id (required), tag (required).
cases-createcases-createC
Creates a new CMMN case
Inputs: name (required), description, parent_case_id, status, +2 more.
cases-deletecases-deleteC
Deletes a case
Inputs: case_id (required).
cases-getcases-getB
Gets a specific case by ID
Inputs: case_id (required).
cases-get_parentcases-get_parentA
Gets the parent case of a sub-case (if any)
Inputs: case_id (required).
cases-listcases-listC
Lists all cases
Inputs: client_id, name_contains, parent_case_id, root_only, +2 more.
cases-list_documentationcases-list_documentationA
Lists all documentation entries for a case. Returns array of {id, text, textFormat, created_at, created_by}.
Inputs: case_id (required).
cases-list_in_workspacecases-list_in_workspaceB
Lists all cases in a workspace
Inputs: workspace_id (required).
cases-list_subcasescases-list_subcasesA
Lists all sub-cases of a parent case (CaseTask pattern)
Inputs: parent_case_id (required).
cases-load_contextcases-load_contextA
Loads a case's full context in ONE call: case metadata, all items with content, and sub-cases recursively. Use this instead of multiple cases-get + items-list + items-get calls when the user says 'load this case into context' or similar.
Inputs: case_id (required), depth, exclude_types, max_content_bytes.
cases-remove_documentationcases-remove_documentationB
Removes a documentation entry from a case by its ID.
Inputs: case_id (required), doc_id (required).
cases-remove_tagcases-remove_tagB
Removes a tag from a case
Inputs: case_id (required), tag (required).
cases-updatecases-updateB
Updates a case name, description, and/or parent case
Inputs: case_id (required), description, name, parent_case_id.
cases-update_statuscases-update_statusA
Updates a case's lifecycle status with CMMN transition validation. Automatically sets timestamps. Valid transitions: active → completed|terminated|suspended|failed, suspended → active|terminated, completed → active|closed, terminated → closed, failed → closed, closed → active.
Inputs: case_id (required), status (required).
clients-createclients-createC
Creates a new client (billing entity)
Inputs: name (required), contact_info, currency, default_rate, +1 more.
clients-deleteclients-deleteC
Deletes a client
Inputs: client_id (required).
clients-getclients-getB
Gets a client by ID
Inputs: client_id (required).
clients-get_by_nameclients-get_by_nameB
Gets a client by name
Inputs: name (required).
clients-listclients-listA
Lists all clients
Inputs: no named parameters.
clients-updateclients-updateC
Updates a client
Inputs: client_id (required), contact_info, currency, default_rate, +2 more.
cmmn-achieve_milestonecmmn-achieve_milestoneA
Achieves a milestone (transitions from available to completed). Sets achieved_at timestamp.
Inputs: milestone_id (required).
cmmn-activate_event_listenercmmn-activate_event_listenerA
Activates an event listener (available -> active). For timers, calculates and schedules the next trigger time.
Inputs: listener_id (required).
cmmn-add_discretionary_itemcmmn-add_discretionary_itemB
Adds a discretionary item to a running case by instantiating from a PlanItemDefinition. The definition must be marked as discretionary.
Inputs: case_id (required), definition_id (required), name, parent_id, +1 more.
cmmn-add_model_to_casecmmn-add_model_to_caseA
Adds items from a Case Plan Model to an existing case. Creates a 'model' container item that holds all model items as children, allowing the entire workflow to be managed as a unit. Returns model_item_id for the container.
Inputs: case_id (required), model_id (required), prefix.
cmmn-check_sentry_conditioncmmn-check_sentry_conditionA
Checks if a single condition is satisfied without updating any sentry. Useful for AI to preview conditions.
Inputs: condition, event, source_ref.
cmmn-claim_complete_with_evidencecmmn-claim_complete_with_evidenceA
Judge-layer task completion (case #1:4264 phase 3). Evaluates the task's acceptance_criteria against the supplied evidence. On satisfied: completes the task. On not_satisfied: returns a judge_rejected halt with per-criterion verdicts. On uncertain (manual criteria present): returns user_review_required (phase 4 will introduce LLM judging). Tasks with no acceptance_criteria fall through to cmmn-complete_task semantics.
Inputs: evidence (required), task_id (required), output_summary, result.
cmmn-claim_stage_complete_with_evidencecmmn-claim_stage_complete_with_evidenceA
Phase 7 — stage analogue of cmmn-claim_complete_with_evidence. Evaluates the stage's acceptance_criteria against evidence; on satisfied, transitions the stage to completed. Otherwise returns a judge_rejected or judge_uncertain halt.
Inputs: evidence (required), stage_id (required), output_summary, result.
cmmn-complete_taskcmmn-complete_taskA
Completes a task. Stores result in task.data.result, sets status to completed, and evaluates sentries. Pre-flight gates: human tasks require confirm_human=true; tasks in 'available' state with unsatisfied entry sentries return 'blocked_by'; tasks with acceptance_criteria require the judge-layer evidence path — call cmmn-claim_complete_with_evidence instead, OR use force=true here to bypass. Use force=true to bypass any gate when the user has explicitly confirmed.
Inputs: task_id (required), confirm_human, force, result.
cmmn-create_event_listenercmmn-create_event_listenerB
Creates a CMMN Event Listener - timer, user, or generic event that can trigger sentries.
Inputs: case_id (required), listener_type (required), name (required), authorized_roles, +9 more.
cmmn-create_milestonecmmn-create_milestoneB
Creates a CMMN Milestone - an achievable target that marks important progress (no work, just achievement).
Inputs: case_id (required), name (required), status (required), description, +2 more.
cmmn-create_planning_tablecmmn-create_planning_tableB
Creates a Planning Table that defines which discretionary items can be added at runtime to a stage or case.
Inputs: case_id (required), name (required), applies_to, applies_to_type, +3 more.
cmmn-create_sentrycmmn-create_sentryB
Creates a CMMN Sentry with optional CMMN 1.1 enhancements (multiple conditions, evaluation modes).
Inputs: case_id (required), name (required), sentry_type (required), condition, +11 more.
cmmn-create_stagecmmn-create_stageC
Creates a CMMN Stage for grouping related plan items (tasks, milestones, other stages).
Inputs: case_id (required), name (required), status (required), auto_complete, +4 more.
cmmn-create_taskcmmn-create_taskC
Creates a CMMN-aligned task with lifecycle state management
Inputs: title (required), case_id (required), status (required), auto_complete_on_sub_case, +30 more.
cmmn-evaluate_sentrycmmn-evaluate_sentryB
Evaluates a sentry's conditions and updates its satisfied status. Returns detailed results for each condition.
Inputs: sentry_id (required), context.
cmmn-execute_stagecmmn-execute_stageA
Executes a stage by traversing its task tree depth-first. Human tasks pause execution. IMPORTANT: When the result contains next_actions with action_required='llm_execute', YOU (the calling LLM) MUST process each prompt immediately. Read the 'prompt' field, execute the work it describes, then call cmmn-complete_task with the task_id and your result. Do NOT just report these back to the user — you are the executor. The stage is not complete until all llm_execute actions have been processed and their tasks completed.
Inputs: stage_id (required), context, human_task_mode, max_depth, +1 more.
cmmn-execute_taskcmmn-execute_taskA
Executes a single task directly. Process/decision/case tasks execute automatically. Human tasks return an error (use update_task_status instead). IMPORTANT: When the result contains action_required='llm_execute' or a 'prompt' field, YOU (the calling LLM) MUST execute the prompt immediately, then call cmmn-complete_task with the task_id and your result. You are the executor — do the work, don't just report it.
Inputs: task_id (required), confirm_human, context, force, +1 more.
cmmn-get_acceptance_criteriacmmn-get_acceptance_criteriaA
Returns the judge-layer acceptance criteria + required_evidence for a task. Each criterion has id, text, verifier. verifier is a registered automated check name (e.g. schema_check, file_exists) or 'manual' (deferred to LLM judge). Backwards-compatible: if only legacy data.exit_criteria strings exist, they're normalized to manual criteria with auto-generated ids. Phase 1 of case #1:4264.
Inputs: task_id (required).
cmmn-get_execution_flowcmmn-get_execution_flowA
Gets the execution flow for a case by traversing OnPart and Guards edges. Returns workflow connections showing which items trigger sentries (OnPart) and which items are guarded by sentries (Guards). Useful for visualizing CMMN workflow dependencies.
Inputs: case_id (required).
cmmn-get_execution_plancmmn-get_execution_planA
Preview execution order for a stage without executing. Shows which items are blocked by sentries and execution dependencies.
Inputs: stage_id (required), include_blocked.
cmmn-get_nextcmmn-get_nextA
Gets the next task to execute in a case workflow. Returns the task with gathered context (from context_sources), prompt, and result_schema. For human tasks, this is what to present to the user. For process tasks, includes execution instructions.
Inputs: case_id (required).
cmmn-get_resume_contextcmmn-get_resume_contextA
Gets context for AI to resume work on a case. Returns active/suspended/enabled tasks, recent completions, and summary.
Inputs: case_id (required).
cmmn-list_added_modelscmmn-list_added_modelsA
Lists all models that have been added to a case (via add_model_to_case or instantiate).
Inputs: case_id (required).
cmmn-list_discretionary_itemscmmn-list_discretionary_itemsB
Lists all discretionary items available for a stage or case. Returns planning tables and their available PlanItemDefinitions.
Inputs: target_id (required), target_type (required).
cmmn-list_itemscmmn-list_itemsA
Lists CMMN plan items for a case. Filter by type (task/stage/milestone/event_listener/sentry) and/or status.
Inputs: case_id (required), status, type.
cmmn-list_sentries_for_itemcmmn-list_sentries_for_itemB
Lists all sentries (entry and exit criteria) that guard an item.
Inputs: item_id (required), sentry_type.
cmmn-mark_sentry_part_satisfiedcmmn-mark_sentry_part_satisfiedB
Marks a specific condition/part of a sentry as satisfied (for onPart evaluation mode).
Inputs: part_index (required), sentry_id (required).
cmmn-mark_sentry_satisfiedcmmn-mark_sentry_satisfiedB
Marks a sentry as satisfied. Sets satisfied=true and records timestamp.
Inputs: sentry_id (required).
cmmn-reset_sentrycmmn-reset_sentryA
Resets a sentry to unsatisfied state, clearing all satisfied_parts. Used when re-evaluation is needed.
Inputs: sentry_id (required).
cmmn-reset_stagecmmn-reset_stageA
Resets a completed stage to active state for re-execution. The stage executor can then re-process its child tasks. Use cascade=true to also reset all downstream items.
Inputs: stage_id (required), cascade.
cmmn-reset_taskcmmn-reset_taskA
Resets a task to allow re-execution. Returns reset_items audit listing every item touched (id, name, type, from_status, to_status). With cascade=true, walks the sentry graph forward and resets all downstream items so cmmn-execute_stage will re-run them rather than skipping completed work. With dry_run=true, returns the same audit without applying. Existing result data and child items are kept.
Inputs: task_id (required), cascade, dry_run, to_status.
cmmn-set_acceptance_criteriacmmn-set_acceptance_criteriaA
Sets the judge-layer acceptance criteria on a task. The runtime judge (case #1:4264 phase 3) will evaluate these when the agent claims completion via cmmn-claim_complete_with_evidence. Workflow authors call this to attach verification recipes — schema_check for SQL migrations, command_runs_clean for build steps, manual for natural-language assertions.
Inputs: criteria (required), task_id (required), required_evidence.
cmmn-terminate_event_listenercmmn-terminate_event_listenerA
Terminates an event listener (cancels any scheduled timer jobs).
Inputs: listener_id (required).
cmmn-trigger_eventcmmn-trigger_eventB
Manually triggers an event listener (marks as completed). Use for user events or testing.
Inputs: listener_id (required), event_data.
cmmn-update_sentrycmmn-update_sentryA
Updates a sentry's fields after creation. If guards_item or source_ref change, the corresponding graph edges (Guards/OnPart) are automatically deleted and recreated.
Inputs: sentry_id (required), condition, condition_type, evaluation_mode, +3 more.
cmmn-update_stage_statuscmmn-update_stage_statusA
Updates a stage's status with CMMN transition validation. Automatically sets timestamps. Valid transitions: available → enabled|disabled, enabled → active|disabled, disabled → enabled, active → completed|terminated|failed|suspended, suspended → active|terminated. Note: Stages with acceptance_criteria require cmmn-claim_stage_complete_with_evidence when transitioning to 'completed' (or pass force=true).
Inputs: stage_id (required), status (required), force.
cmmn-update_task_statuscmmn-update_task_statusA
Updates a task's status with CMMN transition validation. Automatically sets timestamps. Valid transitions: available → enabled|disabled, enabled → active|disabled, disabled → enabled, active → completed|terminated|failed|suspended, suspended → active|terminated. Tip: To complete a task from any startable state (available/enabled/active), use cmmn-complete_task instead — it auto-transitions through intermediate states.
Inputs: status (required), task_id (required).
cmmn-validate_timer_expressioncmmn-validate_timer_expressionA
Validates a timer expression without creating anything. Returns parsed details (seconds for duration, datetime for date, next_trigger for cron) or an error message. Use before creating timer event listeners to catch expression errors early.
Inputs: expression (required), timer_type (required).
cmmn-verify_sentry_edgescmmn-verify_sentry_edgesA
Verifies sentry JSON data matches graph edges. Returns count of matching sentries and any mismatches where source_ref or guards_item JSON fields don't match OnPart/Guards edges. Useful for data integrity checks.
Inputs: case_id (required).
files-find_by_namefiles-find_by_nameA
Finds a file by name within a case. Returns the file item if found, null if not.
Inputs: case_id (required), filename (required).
files-list_versionsfiles-list_versionsA
Lists all versions of a file, including current version info and version history
Inputs: item_id (required).
files-readfiles-readA
Reads file content from storage. Images (PNG, JPEG, GIF, WebP, SVG) are returned as native MCP image content blocks that LLMs can view directly. Non-image files return base64-encoded content by default, or plain text if encoding='text'.
Inputs: item_id (required), encoding.
graph-get_linkedgraph-get_linkedA
Gets all items linked from a vertex. Optionally filter by link_type (e.g., 'context').
Inputs: from_id (required), link_type.
graph-linkgraph-linkA
Creates a LinkedTo edge between any two vertices. Use link_type='context' to link a case as a context source.
Inputs: from_id (required), to_id (required), link_type.
graph-unlinkgraph-unlinkB
Removes a LinkedTo edge between two vertices
Inputs: from_id (required), to_id (required).
invoices-createinvoices-createB
Creates a standalone Invoice vertex with a BillTo edge to the specified client.
Inputs: client_id (required), invoice_number (required), amount, currency, +4 more.
invoices-deleteinvoices-deleteC
Deletes an invoice
Inputs: invoice_id (required).
invoices-getinvoices-getB
Gets an invoice by ID
Inputs: invoice_id (required).
invoices-get_clientinvoices-get_clientB
Gets the client for an invoice via BillTo edge. Returns the Client details.
Inputs: invoice_id (required).
invoices-get_durationsinvoices-get_durationsC
Gets all durations linked to an invoice
Inputs: invoice_id (required).
invoices-get_expensesinvoices-get_expensesA
Gets all expenses linked to an invoice
Inputs: invoice_id (required).
invoices-link_iteminvoices-link_itemB
Links a duration or expense to an invoice
Inputs: invoice_id (required), item_id (required).
invoices-listinvoices-listA
Lists invoices, optionally filtered by client, billing period, or status
Inputs: client_id, month_of, status.
invoices-list_unbilled_durationsinvoices-list_unbilled_durationsB
Lists all durations not yet linked to any invoice
Inputs: case_id.
invoices-list_unbilled_expensesinvoices-list_unbilled_expensesA
Lists all expenses not yet linked to any invoice
Inputs: case_id.
invoices-unlink_iteminvoices-unlink_itemB
Unlinks a duration or expense from an invoice
Inputs: invoice_id (required), item_id (required).
invoices-updateinvoices-updateC
Updates an invoice
Inputs: invoice_id (required), amount, currency, invoice_date, +4 more.
items-add_childitems-add_childB
Adds an existing item as a child of another item
Inputs: child_id (required), parent_id (required).
items-add_documentationitems-add_documentationB
Adds a documentation entry to a CaseFileItem following the CMMN tCmmnElement pattern. Documentation is stored in data.documentation array.
Inputs: item_id (required), text (required), text_format.
items-convert_typeitems-convert_typeA
Converts a CaseFileItem from one type to another. Supported conversions among todo, note, task, and checkbox (todo<->note, todo<->task, note<->task, todo<->checkbox, note<->checkbox). Previous data is preserved for recovery.
Inputs: item_id (required), new_type (required).
items-createitems-createA
Creates a case file item. Set 'type' to choose the item type. Common params: case_id, name/title, parent_id, tags, description. Type-specific: note(content), bookmark(url,title), todo(title,status,priority,due_date), duration(start_time,end_time,billable), expense(description,amount,date), invoice(invoice_number,amount,client_id), file(filename), folder(name), checkbox(label,checked), journal_entry(content,mood), sms(body,to), directory(path), desktop(name).
Inputs: case_id (required), type (required), filename, notes, +35 more.
items-deleteitems-deleteA
Deletes a case file item. If it's a file, also removes the stored file (unless referenced by other items)
Inputs: item_id (required).
items-delete_orphaneditems-delete_orphanedA
[Admin only] Deletes all orphaned CaseFileItems (items with no parent case). Returns count of deleted items.
Inputs: no named parameters.
items-find_by_urlitems-find_by_urlA
Finds bookmark items matching a URL across all cases. Returns all bookmarks with this URL and their case_id. Used by the browser extension to detect if the current page is bookmarked.
Inputs: url (required).
items-getitems-getA
Gets a single case file item by ID
Inputs: item_id (required).
items-get_batchitems-get_batchA
Gets multiple case file items by IDs in a single request. Returns a map of items keyed by ID.
Inputs: ids (required).
items-get_parentitems-get_parentA
Gets the parent of a case file item (if any)
Inputs: item_id (required).
items-listitems-listB
Lists all case file items for a case. Use filters to reduce token usage.
Inputs: case_id (required), exclude_data, limit, name_contains, +1 more.
items-list_childrenitems-list_childrenB
Lists all children of a case file item
Inputs: parent_id (required).
items-list_documentationitems-list_documentationB
Lists all documentation entries for a CaseFileItem. Returns array of {id, text, textFormat, created_at, created_by}.
Inputs: item_id (required).
items-list_orphaneditems-list_orphanedA
[Admin only] Lists all orphaned CaseFileItems (items with no parent case). Use this to find leaked items.
Inputs: no named parameters.
items-moveitems-moveB
Moves an item to a new parent or removes its parent
Inputs: item_id (required), parent_id.
items-move_to_caseitems-move_to_caseA
Moves a CaseFileItem (and all descendants) to a different case. Can move to case root or into a specific folder.
Inputs: case_id (required), item_id (required), parent_id.
items-remove_documentationitems-remove_documentationB
Removes a documentation entry from a CaseFileItem by its ID.
Inputs: doc_id (required), item_id (required).
items-searchitems-searchA
Search CaseFileItems by name across all cases owned by user
Inputs: query (required), limit, type.
items-semantic_searchitems-semantic_searchA
Semantic search across cases and items using natural language. Unlike items-search which matches by name only, this finds conceptually related items even when exact terms don't match. Requires embeddings to be enabled (EMBEDDINGS_ENABLED=true).
Inputs: query (required), entity_types, limit, min_similarity, +1 more.
items-updateitems-updateB
Updates a case file item (name or data)
Inputs: item_id (required), data, name.
items-upload_fileitems-upload_fileA
Uploads a local file to a case in one MCP call. The agent reads the file from the local filesystem (e.g., via Read or base64 < file), passes the encoded bytes, and the server decodes and stores them in FileStore as a real binary file (identical to POST /api/upload). Best for screenshots and small/medium files — base64 inflates payload ~33%, so for files larger than a few MB use the curl path to /api/upload instead. Use parent_id to attach the file as a child of a note or task in one call.
Inputs: case_id (required), content_base64 (required), filename (required), mime_type, +2 more.
messages-archivemessages-archiveA
Archives a message and all its children (ai_work_items). Sets status='archived' on the message and cascades to child items. Archived messages are hidden from the default inbox view but remain accessible via the 'Show archived' filter.
Inputs: message_id (required).
messages-broadcastmessages-broadcastA
Fan-out variant of messages-send. Resolves to_capability against the recipient's online agents and dispatches one message per match. Returns {success: true, count: N, message_ids: [...]} or an error if no agents match.
Inputs: content (required), to_capability (required), context_case_id, priority, +4 more.
messages-listmessages-listC
Lists messages received by the current user. Defaults to unread messages in the personal Inbox case. Filtering on status/from_user_id/reply_to_message_id happens in Elixir since the message data is stored as a JSON string.
Inputs: case_id, from_user_id, limit, reply_to_message_id, +1 more.
messages-readmessages-readA
Fetches the full content of a single message and marks it as read (unless mark_read=false). Returns 404 if the item is not a message or the user does not own its case.
Inputs: message_id (required), mark_read.
messages-sendmessages-sendA
Sends a direct message to another user. Stored as a CaseFileItem of type 'message' in the recipient's inbox case (or a specified context case). Returns the message_id.
Inputs: content (required), context_case_id, priority, reply_to_message_id, +5 more.
messages-unarchivemessages-unarchiveA
Restores an archived message by setting its status back to 'read'. Does not affect children.
Inputs: message_id (required).
meta-desire_pathsmeta-desire_pathsA
Shows tools that agents tried to call but don't exist — desire paths that reveal feature requests from the field. Queries historical data from ArcadeDB.
Inputs: limit, min_count, since.
meta-embeddings_backfillmeta-embeddings_backfillA
Backfills semantic search embeddings for all existing cases and items. Run this once after enabling embeddings to index existing content. Returns the count of items queued for embedding.
Inputs: no named parameters.
meta-embeddings_backfill_allmeta-embeddings_backfill_allA
Backfills semantic search embeddings for ALL users that have cases. Returns per-user counts and total items queued.
Inputs: no named parameters.
meta-embeddings_statsmeta-embeddings_statsA
Returns statistics about semantic search embeddings: total count, counts by type, and whether embeddings are enabled.
Inputs: no named parameters.
meta-flush_logsmeta-flush_logsA
Forces an immediate flush of in-memory tool call logs to ArcadeDB for permanent storage.
Inputs: no named parameters.
meta-get_playbookmeta-get_playbookA
Returns a detailed playbook/skill guide. Available: cmmn-execution (comprehensive), stage-execution, ai-work-items, session-lifecycle, getting-started, file-uploads, model-authoring, agent-messaging, orchestration.
Inputs: name (required).
meta-get_startedmeta-get_startedA
Returns a getting-started guide for LLMs using CaseMgr. Includes CMMN concepts, common workflows, and tips. Call this first if you're unfamiliar with CaseMgr.
Inputs: no named parameters.
meta-preview_emailmeta-preview_emailA
Admin only. Sends an email template preview to a whitelisted test address. Use for iterating on email designs without creating real users.
Inputs: template (required), recipient, variant.
meta-tool_statsmeta-tool_statsA
Real-time MCP tool call statistics from the current period. Shows call counts, error rates, and average durations aggregated by tool and status.
Inputs: limit, status.
models-add_definitionmodels-add_definitionA
Adds a plan item definition to a draft model. Use parent_definition_id to nest items inside stages. For sentries, use guards_definition_id and source_definition_id. For event listeners, use triggers_sentry_definition_id.
Inputs: definition_type (required), model_id (required), name (required), discretionary, +10 more.
models-archivemodels-archiveA
Archives a Case Plan Model. Archived models cannot be instantiated.
Inputs: model_id (required).
models-createmodels-createB
Creates a new Case Plan Model (CMMN 1.1). Models start in 'draft' status.
Inputs: name (required), description, version.
models-create_from_stagemodels-create_from_stageA
Creates a new draft model from an existing stage. Extracts CMMN items (tasks, milestones, sentries, event listeners) as model definitions. Non-CMMN items (notes, todos) are excluded. Sentry connections are preserved. The model has no references to the source case.
Inputs: stage_id (required), description, name.
models-create_new_versionmodels-create_new_versionA
Creates a new draft version of a model, copying all definitions. Auto-increments minor version (1.0 → 1.1) unless a version override is provided.
Inputs: model_id (required), version.
models-deletemodels-deleteA
Deletes a draft Case Plan Model (published/archived models cannot be deleted)
Inputs: model_id (required).
models-delete_definitionmodels-delete_definitionA
Deletes a plan item definition from a draft model. Alias for models-remove_definition — provided for naming consistency with the rest of the *-delete_* tools (models-delete, items-delete, cases-delete).
Inputs: definition_id (required), model_id (required), dry_run, repoint_inbound.
models-getmodels-getB
Gets a Case Plan Model by ID
Inputs: model_id (required).
models-get_for_casemodels-get_for_caseA
Gets the model a case was instantiated from (if any)
Inputs: case_id (required).
models-instantiatemodels-instantiateA
Creates a new Case from a published model, with items from plan item definitions. Pass workspace_id to make the case visible in a workspace.
Inputs: model_id (required), name, parent_case_id, status, +2 more.
models-listmodels-listA
Lists Case Plan Models (your own, optionally including public models from other users)
Inputs: include_public, status.
models-list_definitionsmodels-list_definitionsB
Lists all plan item definitions for a model
Inputs: model_id (required).
models-list_instancesmodels-list_instancesB
Lists all cases instantiated from a model
Inputs: model_id (required).
models-publishmodels-publishA
Publishes a Case Plan Model (draft -> published). If other published versions of the same model exist, they are listed in the response as prior_published_versions with a hint to archive them. Pass archive_version_ids to publish AND archive in one call.
Inputs: model_id (required), archive_version_ids.
models-remove_definitionmodels-remove_definitionA
Removes a plan item definition from a draft model. For task definitions, automatically bridges the inbound sentries to the outbound targets so the workflow chain stays intact (the dropped-task case). For non-task definitions (sentries, milestones, stages, event_listeners), cascades to any dependent sentry. Also exposed as models-delete_definition (alias).
Inputs: definition_id (required), model_id (required), dry_run, repoint_inbound.
models-set_visibilitymodels-set_visibilityA
Sets model visibility. Public models can be viewed and instantiated by any user. Only the owner can change visibility.
Inputs: model_id (required), visibility (required).
models-updatemodels-updateA
Updates a Case Plan Model (only draft models can have name/version/description updated)
Inputs: model_id (required), description, name, version.
models-update_definitionmodels-update_definitionA
Updates a plan item definition in a draft model. Merges template_data with existing values. Can also change the top-level sequence_order and parent_definition_id without touching template_data. Use to change task_type, description, status, ordering, or parent placement.
Inputs: definition_id (required), model_id (required), name, parent_definition_id, +2 more.
models-validatemodels-validateA
Validates a model for errors before publishing. Checks: orphan references, event type mismatches, duplicate sentries, stages without milestones, sequence order conflicts, ungated tasks, empty stages, disconnected milestones. Returns {errors, warnings, info} report. Call this before models-publish.
Inputs: model_id (required).
preferences-getpreferences-getA
Gets the current user's preferences. Pass path and machine_id to auto-resolve the worktree case as current_case_id (falls back to global current_case_id if no worktree match). Response includes case_source: 'worktree' or 'global'.
Inputs: machine_id, path.
preferences-list_pinned_casespreferences-list_pinned_casesA
Lists all cases pinned by the current user across all workspaces
Inputs: no named parameters.
preferences-set_brain_casepreferences-set_brain_caseA
Sets or clears the user's brain case for thought capture. When set, inbound SMS from the user's phone number is automatically stored as an item in this case and embedded for semantic search. Also used by meta-quick_capture for frictionless note capture.
Inputs: case_id.
preferences-set_current_casepreferences-set_current_caseA
Sets or clears the user's current active case. Omit case_id to clear.
Inputs: case_id.
preferences-toggle_pinpreferences-toggle_pinB
Pins or unpins a case for quick access in the Pinned Cases view.
Inputs: case_id (required), pinned (required).
preferences-updatepreferences-updateC
Updates the current user's preferences
Inputs: default_workspace_id, locale, settings, theme, +1 more.
process-create_definitionprocess-create_definitionA
Creates a reusable process definition for automated workflows. Use with process tasks (task_type='process').
Inputs: action_type (required), case_id (required), name (required), agent_config, +9 more.
process-create_taskprocess-create_taskA
Creates a process task linked to a definition. When activated (status=active), the process will execute automatically.
Inputs: case_id (required), process_definition_id (required), title (required), description, +3 more.
process-delete_definitionprocess-delete_definitionA
Deletes a process definition. Note: This orphans any tasks referencing this definition.
Inputs: definition_id (required).
process-execute_taskprocess-execute_taskA
Manually executes a process task. The task must have task_type='process' and a process_definition_id. Returns execution result.
Inputs: task_id (required).
process-get_definitionprocess-get_definitionA
Gets a process definition by ID
Inputs: definition_id (required).
process-list_definitionsprocess-list_definitionsC
Lists all process definitions in a case
Inputs: case_id (required), action_type.
process-update_definitionprocess-update_definitionC
Updates a process definition
Inputs: definition_id (required), agent_config, automation_config, completion_steps, +8 more.
sms-configuredsms-configuredA
Checks if Twilio SMS credentials are configured.
Inputs: no named parameters.
sms-list_messagessms-list_messagesC
Lists all SMS messages (sent and received) stored as CaseFileItems for a case.
Inputs: case_id (required).
sms-list_templatessms-list_templatesA
Lists all available SMS templates with their descriptions and required parameters.
Inputs: no named parameters.
sms-sendsms-sendA
Sends an SMS message to a phone number. Requires the 'sms' feature to be enabled. Message is queued for async delivery.
Inputs: body (required), to (required), case_id.
sms-send_templatesms-send_templateB
Sends a templated SMS message. Renders the template with provided params and sends via Twilio.
Inputs: params (required), template (required), to (required), case_id.
tasks-check_todostasks-check_todosA
Checks if all todos under a task are complete. Returns completion status and counts.
Inputs: task_id (required).
tasks-create_todotasks-create_todoB
Creates a todo as a subtask of a CMMN task. The todo is linked as a child via ParentOf edge.
Inputs: case_id (required), task_id (required), title (required), assigned_to, +4 more.
tasks-list_case_todostasks-list_case_todosA
Lists all todos for a case, sorted by priority and due date
Inputs: case_id (required), status.
tasks-list_todostasks-list_todosA
Lists all todos that are subtasks of a CMMN task (linked via ParentOf edge)
Inputs: task_id (required).
tasks-update_todo_statustasks-update_todo_statusA
Updates a todo item's status with automatic timestamp handling
Inputs: status (required), todo_id (required).
tokens-createtokens-createA
Creates a new API token for the current user. Returns the full token ONCE - save it, it cannot be retrieved again.
Inputs: name (required).
tokens-deletetokens-deleteB
Revokes/deletes an API token for the current user
Inputs: token_id (required).
tokens-listtokens-listA
Lists all API tokens for the current user (only shows token prefix, not full token)
Inputs: no named parameters.
users-disableusers-disableA
Disables a user account, logging them out (admin only)
Inputs: user_id (required).
users-enableusers-enableA
Re-enables a disabled user account (admin only)
Inputs: user_id (required).
users-getusers-getA
Gets a user by ID (admin only)
Inputs: user_id (required).
users-get_currentusers-get_currentA
Gets information about the currently authenticated user
Inputs: no named parameters.
users-get_featuresusers-get_featuresA
Gets the enabled features for a user (admin only)
Inputs: user_id (required).
users-listusers-listA
Lists all users in the system (admin only)
Inputs: no named parameters.
users-set_featuresusers-set_featuresA
Sets the enabled features for a user (admin only). Replaces the entire features list.
Inputs: features (required), user_id (required).
users-set_productsusers-set_productsA
Sets the enabled products for a user (admin only). Products control which MCP tool namespaces are visible. User must reconnect MCP for changes to take effect.
Inputs: products (required), user_id (required).
users-update_profileusers-update_profileA
Updates the current user's profile (phone, email). At least one field must be provided.
Inputs: email, phone.
workspaces-add_caseworkspaces-add_caseB
Adds a case to a workspace (cases can be in multiple workspaces)
Inputs: case_id (required), workspace_id (required).
workspaces-createworkspaces-createB
Creates a new workspace for grouping cases
Inputs: name (required), color, description, icon, +1 more.
workspaces-deleteworkspaces-deleteB
Deletes a workspace
Inputs: workspace_id (required).
workspaces-getworkspaces-getB
Gets a workspace by ID
Inputs: workspace_id (required).
workspaces-get_defaultworkspaces-get_defaultA
Gets the user's default workspace
Inputs: no named parameters.
workspaces-listworkspaces-listA
Lists all workspaces for the current user
Inputs: no named parameters.
workspaces-list_for_caseworkspaces-list_for_caseB
Lists all workspaces that contain a case
Inputs: case_id (required).
workspaces-remove_caseworkspaces-remove_caseB
Removes a case from a workspace
Inputs: case_id (required), workspace_id (required).
workspaces-set_defaultworkspaces-set_defaultA
Sets a workspace as the user's default (shown on login)
Inputs: workspace_id (required).
workspaces-updateworkspaces-updateC
Updates a workspace
Inputs: workspace_id (required), color, description, icon, +1 more.
worktrees-activateworktrees-activateA
Activates a worktree by setting its linked case as the current case. Use this when entering a worktree to automatically switch context to its associated case.
Inputs: worktree_id (required).
worktrees-assign_to_caseworktrees-assign_to_caseB
Associates a worktree with a case (creates HasWorktree edge)
Inputs: case_id (required), worktree_id (required).
worktrees-createworktrees-createB
Creates a new worktree entry for git worktree tracking across machines
Inputs: machine_id (required), name (required), path (required), alias, +7 more.
worktrees-deleteworktrees-deleteB
Deletes a worktree
Inputs: worktree_id (required).
worktrees-getworktrees-getA
Gets a worktree by ID, alias (fuzzy), or path. Provide one of: id, alias+machine_id, or path+machine_id
Inputs: alias, machine_id, path, worktree_id.
worktrees-get_caseworktrees-get_caseA
Gets the case associated with a worktree (via HasWorktree edge)
Inputs: worktree_id (required).
worktrees-listworktrees-listA
Lists all worktrees, optionally filtered by machine, remote origin, or branch
Inputs: branch, machine_id, remote_origin.
worktrees-list_for_caseworktrees-list_for_caseC
Lists all worktrees associated with a case
Inputs: case_id (required).
worktrees-unassign_from_caseworktrees-unassign_from_caseA
Removes the association between a worktree and a case (deletes HasWorktree edge)
Inputs: case_id (required), worktree_id (required).
worktrees-updateworktrees-updateB
Updates a worktree's mutable fields
Inputs: worktree_id (required), alias, branch, head, +5 more.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 205 tools
Namespace prefixes help distinguish tools, but overlapping concepts across namespaces — e.g., 'task'/'todo' appears in items-, cmmn-, and tasks- with different semantics, and multiple sentry-evaluation tools (cmmn-check_sentry_condition, cmmn-evaluate_sentry, cmmn-mark_sentry_satisfied) require careful reading. Detailed descriptions mitigate this, but the volume of 205 tools makes selection error-prone.
Tool names consistently follow a `namespace-verb_noun` pattern with snake_case, which is highly predictable. Minor deviations exist: `meta-desire_paths` and `meta-tool_stats` are noun phrases, `calendar-get_or_create` uses a compound verb, and `models-delete_definition` is an alias of `models-remove_definition`. These are small exceptions in a largely uniform convention.
With 205 tools across 20+ namespaces, this is an extremely large surface for any MCP server. Even though the domain is broad, the sheer number overwhelms agents and violates the typical well-scoped server expectation. Many tools could be consolidated or exposed lazily.
The tool set provides CRUD and lifecycle coverage for nearly all entities in the domain: cases, items, models, clients, invoices, users, workspaces, worktrees, and more. Minor gaps exist — e.g., no direct message delete (only archive), no calendar event management, and documentation entries lack an update operation — but agents can work around these.
Maintenance
Related MCP Connectors
Work management where AI agents are first-class members: tasks, projects, memory over hosted MCP
Shared memory for connected AI tools. Projects, rules and skills over MCP. OAuth or API key.
Shared memory for your AI tools: tasks, notes and context every tool can read and write over MCP.
MCP server for your apps' tools and custom tools, plus hosted AI agents and approval-gated workflows
Related MCP Servers
- AlicenseNot gradedqualityBmaintenanceEnd-to-end agent-managed company brain. Humans and any MCP agent co-author living docs (Markdown + extensions), 40+ visual diagrams (Mermaid, BPMN, D2, PlantUML, ELK, Excalidraw), plans, and a self-learning Knowledge Graph. 163 tools across 16 categories. Auth: OAuth 2.1 or API key. Lean, secure, affordable — from individuals to enterprise.MIT
- AlicenseNot gradedqualityDmaintenanceEnables AI assistants and teams to collaborate in a shared workspace by building and querying tables, writing documents, uploading files, and publishing live dashboards through any MCP-compatible client.2MIT
- FlicenseNot gradedqualityDmaintenanceShared memory, skills and agents for your team — Claude, Cursor and other MCP clients start every session with how your firm actually works. Versioned artifacts with a wiki-link graph, remote Streamable HTTP with OAuth 2.1 sign-in.-
- FlicenseNot gradedqualityDmaintenanceProvides a hosted remote MCP server that lets compatible AI clients retrieve and save durable context, checkpoints, operating rules, project notes, and permission-controlled templates for continuity across ChatGPT, Claude, IDEs, coding agents, and other AI tools. It enables token-efficient startup and resume handoffs so users can move projects between tools without rebuilding detailed prompts each session.1-