Skip to main content
Glama

genskill-mcp

An MCP server that lets any AI agent build on GenLayer: search the docs, inspect contracts and transactions over RPC, and scaffold, lint, and test Intelligent Contracts.

npm CI License: MIT Node MCP

genskill-mcp is a Model Context Protocol server for MCP-compatible clients (Claude Code, Cursor, VS Code, Gemini CLI, Codex, and remote MCP clients over HTTP). It combines searchable GenLayer documentation, live protocol/RPC inspection of contracts and transactions, and a contract authoring toolkit (scaffold, lint, and test Intelligent Contracts), so an agent can go from zero to a deployable contract without leaving its editor.

This project does not sign transactions or manage private keys. It supports live node inspection and contract interaction through GenLayer JSON-RPC methods such as gen_call, gen_getContractState, gen_getContractCode, gen_getContractSchema, gen_getTransactionStatus, and gen_getTransactionReceipt.

Install

Use

Command / URL

Local (stdio)

npx -y genskill-mcp

Hosted (remote)

https://genskill-mcp.vercel.app/mcp

Per-client setup (Claude Code, Cursor, VS Code, Gemini, Codex, remote) is in the quickstarts below.

Related MCP server: GenLayer MCP Server

What this server does

The server loads the official GenLayer docs bundle:

It parses that bundle into sections and exposes:

  • searchable MCP tools

  • a browsable docs index resource

  • individual section resources

  • live GenLayer JSON-RPC tools for contract and transaction inspection

  • a browsable RPC configuration resource

  • contract authoring tools: scaffold a starter contract and lint it before deploy

  • guided MCP prompts for writing, testing, and debugging contracts

This repository supports two transports:

  • stdio for local CLI tools such as Claude Code, Codex, Cursor, VS Code, and Gemini CLI

  • Streamable HTTP for deployed remote MCP usage

Build a contract (authoring tools, new in 2.2)

If you are new to GenLayer, this is the fast path from zero to a deployable contract. It does not just point you at docs, it gives you working code and checks it.

  • genlayer_scaffold_contract generates a working starter Intelligent Contract for a template (storage, llm-judge, web-oracle, token). The output already has the runner header pinned and avoids the common GenVM deploy-killers, so it deploys as-is.

  • genlayer_lint_contract runs static pre-deploy checks on contract source. It catches the mistakes that make a deploy finalize with a bare invalid_contract (no stack trace): a comment line directly under the runner header, an unpinned or :test / :latest runner, a missing gl.Contract class, forbidden sandbox imports (os, sys, subprocess, random, ...), and GenVM Python-subset issues such as for loops, sorted, .sort, lambda, and list / dict storage fields.

  • genlayer_scaffold_test generates a fast direct-mode test (genlayer-test) for a template, using the real fixtures (direct_vm, direct_deploy, direct_alice) and mocking web/LLM calls so tests stay deterministic.

Three guided prompts wrap the workflow for any MCP client:

  • genlayer_write_contract scaffold, decide what needs consensus, then lint.

  • genlayer_test_contract direct-mode then integration testing.

  • genlayer_debug_deploy the checklist for a deploy that errored on-chain.

And a single always-available resource, genlayer://guide/contract-rules, is an authoritative cheat sheet (skeleton, storage types, web/LLM/equivalence APIs, the GenVM Python-subset rules, and the deploy/finalize/debug checklist) that any connecting agent can read to get contract writing and debugging right. It tracks the latest GenLayer docs and the official write-contract / direct-tests skills.

Typical loop: scaffold, edit, genlayer_lint_contract until it is clean, deploy, then inspect it live with genlayer_get_contract_snapshot.

RPC configuration

Live protocol tools use the configured GenLayer RPC endpoint.

  • GENLAYER_RPC_URL: JSON-RPC endpoint to use

  • GENLAYER_RPC_TIMEOUT_MS: timeout for live RPC requests

If GENLAYER_RPC_URL is not set, the server defaults to:

https://studio.genlayer.com/api

Quickstart for Claude Code

claude mcp add --transport stdio genskill -- npx -y genskill-mcp

Then start Claude Code:

claude

Inside Claude Code, run:

/mcp

The genskill server and its tool endpoints should be listed.

Quickstart for Cursor (per-project)

Add this to .cursor/mcp.json:

{
  "mcpServers": {
    "genskill": {
      "command": "npx",
      "args": ["-y", "genskill-mcp"]
    }
  }
}
TIP

If a Cursor version does not recognizemcpServers, use mcp_servers as the top-level key instead.

Quickstart for VS Code (per-workspace)

Add this to .vscode/mcp.json:

{
  "servers": {
    "genskill": {
      "type": "stdio",
      "command": "npx",
      "args": ["-y", "genskill-mcp"]
    }
  },
  "inputs": []
}

Quickstart for Gemini CLI

Add the MCP server globally:

gemini mcp add --scope user genskill npx -y genskill-mcp

Confirm it is registered:

gemini mcp list

Quickstart for Codex

Add the MCP server with the Codex CLI:

codex mcp add genskill -- npx -y genskill-mcp

Confirm it is registered:

codex mcp list

Alternatively, add this to a Codex MCP config:

[mcp_servers.genskill]
command = "npx"
args = ["-y", "genskill-mcp"]

Restart Codex if needed so it reloads the MCP config.

Quickstart from source

For source-based usage instead of installing from npm:

  1. Clone the repo:

    git clone https://github.com/Jr-kenny/genskill-mcp
    cd genskill-mcp
  2. Install dependencies and build:

    npm install
    npm run build
  3. Run the local entrypoint:

    node /absolute/path/to/genskill-mcp/dist/cli.js

Substitute that node .../dist/cli.js command in any MCP client config for source-based usage over npx.

Quickstart for remote MCP clients

Remote MCP clients can use the public HTTP endpoint without an account. Anonymous clients receive the stateless documentation, authoring, planning, and protocol tools. Persisted workflow sessions require an optional tenant bearer token.

This is not Claude-specific. It is the deployed transport for any client that can connect to remote MCP servers over Streamable HTTP, including chat-style AI apps where that capability is available.

  1. Deploy this repository to Vercel. No authentication configuration is required for public stateless access.

  2. To enable private workflow sessions for known tenants, configure credentials as a JSON object. Each key is a stable tenant ID and each value is a random bearer token containing at least 32 characters:

    GENSKILL_MCP_TENANT_TOKENS='{"team-a":"replace-with-a-random-32-character-token"}'

    Anonymous users never see workflow session tools or resources, even when tenant credentials are configured.

  3. The deployed MCP endpoint is:

    https://genskill-mcp.vercel.app/mcp
  4. The deployed service also exposes:

    • GET / for a small server info response

    • GET /health for health checks

    • POST /mcp for MCP Streamable HTTP requests

  5. Add the deployed MCP URL in any remote MCP client. Known tenants can optionally send their token to unlock their isolated workflow sessions:

    Authorization: Bearer replace-with-a-random-32-character-token

Use:

https://genskill-mcp.vercel.app/mcp

This applies to Claude-hosted integrations, ChatGPT-style apps, and other remote AI clients where MCP server URLs are supported.

For local HTTP testing, run:

npm install
npm run build
npm run start:http

It listens on port 3000 by default. Set the PORT environment variable to change it.

HTTP workflow session tools are hidden from anonymous users. Authenticated sessions are stored in separate hashed directories for each tenant. Set GENSKILL_MCP_SESSION_DIR to choose the session root. Vercel defaults to temporary function storage, so use a durable tenant-aware backing store if sessions must survive instance recycling.

Local filesystem access is available only over the stdio transport. HTTP servers don’t register genlayer_load_contract_artifact, and all other HTTP tools reject contractPath. Remote clients can provide base64-encoded code or a deployed contract address.

Vercel deployment

This repository is configured for Vercel through vercel.json and the Web-standard API functions in api/.

Public MCP endpoint on Vercel:

https://genskill-mcp.vercel.app/mcp

Useful checks:

  • https://genskill-mcp.vercel.app/

  • https://genskill-mcp.vercel.app/health

  • POST https://genskill-mcp.vercel.app/mcp from an anonymous or authenticated MCP client

Notes:

  • /mcp is routed to the Vercel function at /api/mcp.mjs.

  • Anonymous /mcp requests expose only stateless tools. A supplied bearer token must be valid, and invalid tokens return 401.

  • The Vercel endpoint uses the MCP SDK Web-standard Streamable HTTP transport in stateless mode.

  • POST requests return JSON responses where possible, which is friendlier for serverless hosting than holding long SSE streams open.

  • If the docs update while the service is already deployed, call genlayer_refresh_docs from an MCP client or redeploy the project.

Tool endpoints

High-level orchestration tools now use a canonical machine-readable response shape with:

  • kind

  • summary

  • current_state

  • blockers

  • next_actions

  • fallbacks

  • data

The autopilot and capability surfaces are capability-aware: they should prefer only actions supported by the currently configured endpoint and explicitly downgrade unsupported debug or ops paths into fallbacks.

Live protocol tools

  1. genlayer_list_networks Lists documented GenLayer network presets, chain IDs, and RPC URLs.

  2. genlayer_start_workflow_session Creates a persisted workflow session from a generated contract workflow plan.

  3. genlayer_list_workflow_sessions Lists persisted workflow sessions ordered by most recently updated.

  4. genlayer_get_workflow_session Reads a persisted workflow session by id.

  5. genlayer_update_workflow_step Marks a workflow session step as completed or pending.

  6. genlayer_autopilot_brief Generates a single operator-grade brief with endpoint capabilities, contract context, workflow plans, handoff steps, and relevant docs.

  7. genlayer_load_contract_artifact Loads a local contract artifact or bytecode file and returns base64 plus file metadata. This tool is exposed only over the local stdio transport.

  8. genlayer_probe_endpoint_capabilities Probes which HTTP and RPC surfaces are actually exposed on the configured GenLayer deployment.

  9. genlayer_generate_agent_handoff Generates an explicit ordered handoff bundle so weaker agents know exactly which genskill-mcp tools to call next.

  10. genlayer_get_contract_interface Normalizes a contract schema into constructor, view methods, and write methods.

  11. genlayer_plan_contract_action Builds a schema-validated execution plan for deploy, read, or write actions.

  12. genlayer_plan_contract_workflow Builds a multi-phase contract workflow covering deploy, wait, snapshot, interaction, and diagnosis.

  13. genlayer_run_transaction_report Orchestrates waiting, inspection, status explanation, optional trace lookup, and optional contract snapshot into one report.

  14. genlayer_run_contract_report Orchestrates network context, contract snapshot, interface, workflow, and default plans into one report.

  15. genlayer_generate_typescript_workflow Generates GenLayerJS deploy/read/write snippets from a contract schema or deployed contract.

  16. genlayer_generate_contract_playbook Generates a schema-aware deployment and interaction playbook for a contract.

  17. genlayer_node_health Calls the configured GenLayer node HTTP GET /health endpoint.

  18. genlayer_network_status Returns a combined live snapshot of node health, chain id, block height, sync status, and optional debug ping status.

  19. genlayer_balance Calls the configured GenLayer HTTP GET /balance endpoint for the node operator.

  20. genlayer_eth_get_balance Calls eth_getBalance through the configured GenLayer RPC endpoint.

  21. genlayer_raw_rpc Calls gen_*, eth_*, zks_*, or zksync_* methods directly against the configured endpoint.

  22. genlayer_trace_transaction Calls gen_dbg_traceTransaction when the target node exposes debug methods.

  23. genlayer_metrics Fetches Prometheus-style metrics from the configured HTTP GET /metrics endpoint.

  24. genlayer_submit_raw_transaction Submits a signed raw transaction through eth_sendRawTransaction.

  25. genlayer_inspect_transaction Combines gen_getTransactionStatus, gen_getTransactionReceipt, and eth_getTransactionByHash into one response.

  26. genlayer_wait_for_transaction Polls transaction status until accepted or finalized.

  27. genlayer_explain_transaction_status Interprets transaction status into finality phase, appealability, and next-step guidance.

  28. genlayer_call_contract Executes gen_call for read, write-simulation, or deploy-simulation requests.

  29. genlayer_get_contract_schema Calls gen_getContractSchema for base64-encoded contract code.

  30. genlayer_get_contract_state Calls gen_getContractState for a deployed contract.

  31. genlayer_get_contract_code Calls gen_getContractCode for a deployed contract.

  32. genlayer_get_contract_snapshot Fetches state, deployed code, and derived schema in one call.

  33. genlayer_get_transaction_status Calls gen_getTransactionStatus for lightweight transaction polling.

  34. genlayer_get_transaction_receipt Calls gen_getTransactionReceipt for full processed transaction data.

  35. genlayer_syncing Calls gen_syncing on the configured endpoint.

Documentation tools

  1. genlayer_search_docs Searches the documentation bundle and returns ranked matches with snippets.

  2. genlayer_refresh_docs Force-refreshes the cached GenLayer documentation bundle from the configured source. Use this after the GenLayer team ships new docs, for example new Studio GEN, payable contract, Faucet, Tip Jar, or MetaMask updates.

  3. genlayer_read_doc Reads a section by slug, path, title, or fuzzy query.

  4. genlayer_get_doc_by_slug Reads a section by exact slug, path, docs URL, or resource URI.

  5. genlayer_search_examples Searches example-heavy sections that contain commands, code blocks, SDK snippets, or config examples.

  6. genlayer_get_related_docs Finds related documentation pages based on section path, title, and neighborhood in the docs tree.

  7. genlayer_list_topics Lists top-level GenLayer documentation topics with counts and example pages.

  8. genlayer_list_sections Lists available parsed documentation sections.

Resources

  1. genlayer://protocol/networks Documented GenLayer network presets, RPC URLs, and chain IDs.

  2. genlayer://workflow/sessions List of persisted workflow sessions. HTTP clients see only their authenticated tenant’s sessions.

  3. genlayer://workflow/session/{id} Persisted workflow session with step completion state, scoped to the authenticated HTTP tenant.

  4. genlayer://workflow/autopilot Single operator-grade brief for the configured endpoint and current contract context.

  5. genlayer://protocol/rpc-config JSON document showing the configured RPC endpoint, timeout, and supported helper methods.

  6. genlayer://protocol/capabilities Probed endpoint capabilities showing which HTTP and RPC surfaces are actually exposed.

  7. genlayer://protocol/transaction/{txId} Combined transaction inspection resource for a specific transaction hash.

  8. genlayer://protocol/transaction/{txId}/report Composed transaction report with status interpretation and optional trace data when exposed.

  9. genlayer://protocol/contract/{address}/state Current accepted-state snapshot for a specific deployed contract.

  10. genlayer://protocol/contract/{address}/snapshot Combined state, code, and schema snapshot for a specific deployed contract.

  11. genlayer://protocol/contract/{address}/playbook Schema-aware deployment and interaction playbook for a deployed contract.

  12. genlayer://protocol/contract/{address}/report Composed contract report with network context, snapshot, interface, workflow, and default plans.

  13. genlayer://protocol/contract/{address}/plans Default workflow plus default read/write action plans for a deployed contract.

  14. genlayer://protocol/contract/{address}/method/{method}/plan/{action} Default schema-validated plan for a specific read or write method.

  15. genlayer://docs/index JSON index of all parsed sections.

  16. genlayer://docs/section/{slug} Individual documentation sections as read-only resources.

Project structure

File/Folder

Purpose

src/index.ts

The server logic: loads docs, parses sections, registers tools and resources

src/cli.ts

Entry point that starts the stdio MCP server

src/genlayerAuthoring.ts

Contract scaffolder + linter + test scaffolder (pure, unit-tested)

src/authoringTools.ts

Registers the authoring tools and the dev-workflow prompts

src/mcpResponses.ts

Shared canonical response envelope helpers

src/genlayerDocs.ts

Docs loading, caching, parsing, search, and formatting helpers

src/genlayerRpc.ts

GenLayer RPC, HTTP ops endpoints, network presets, and transaction helpers

src/genlayerContractToolkit.ts

Schema normalization and GenLayerJS workflow/playbook generation

src/genlayerArtifacts.ts

Local artifact loading and hashing for artifact-driven GenLayer workflows

src/genlayerWorkflowSessions.ts

Persisted workflow session state and session formatting helpers

api/

Vercel API functions for the remote MCP, health, and root endpoints

test/

Unit tests for the authoring module (npm test, built-in node:test)

dist/

Compiled JavaScript output generated by npm run build

package.json

Dependencies, scripts, package metadata, and CLI registration

tsconfig.json

TypeScript compiler settings

vercel.json

Vercel routing and function configuration

README.md

Usage and client setup

How it's built

This MCP server is a lightweight TypeScript implementation built on the official MCP SDK.

Core components

  • Built on @modelcontextprotocol/sdk

  • Uses StdioServerTransport for local MCP clients

  • Uses zod to validate tool arguments

  • Fetches the GenLayer docs bundle from the official docs site

  • Parses the bundle into section-level resources based on each # path/to/file.mdx boundary

  • Uses a simple deterministic ranking function over titles, slugs, paths, and body text

Configuration

Optional environment variables:

  • GENLAYER_DOCS_URL: alternate source URL or local file path for the docs bundle

  • GENLAYER_DOCS_CACHE_FILE: cache location for the downloaded bundle

  • GENLAYER_DOCS_REFRESH_HOURS: cache freshness window in hours, default 24

  • GENLAYER_DOCS_TIMEOUT_MS: HTTP timeout in milliseconds, default 15000

The server also refreshes stale in-memory snapshots automatically. For long-running HTTP deployments, call genlayer_refresh_docs when the GenLayer team ships new documentation and an immediate update is needed without restarting the service.

Local development

This section is only for working on the MCP server itself.

npm install
npm run build
npm run check
npm start

npm run check verifies that the server can fetch and parse the live GenLayer docs bundle.

Releasing (maintainers)

Releases are published to npm automatically by .github/workflows/release.yml when a v* tag is pushed. One-time setup: add an npm automation token as the NPM_TOKEN repository secret (gh secret set NPM_TOKEN).

To cut a release:

npm version minor        # bumps package.json, commits, creates the tag vX.Y.Z
git push --follow-tags   # pushes the commit + tag; CI builds, tests, publishes

The workflow checks the tag matches package.json, runs the tests, and publishes with npm provenance. Once published, consumers update with:

npx -y genskill-mcp@latest      # always newest
npm update -g genskill-mcp      # if installed globally

Remote (Vercel) users need no action — the hosted endpoint auto-deploys from main; clients pick up changes on reconnect.

Credits

Built and maintained by:

Available Tools

46 tools
genlayer_autopilot_briefGenerate GenLayer Autopilot BriefA
Read-onlyIdempotent

Generate a single high-context operator brief that combines endpoint capabilities, contract context, workflow plans, weak-agent handoff steps, and relevant docs.

Inputs: code, goal, address, contractPath.

genlayer_balanceGet GenLayer BalanceA
Read-onlyIdempotent

Fetch the node operator balance from the configured GenLayer HTTP /balance endpoint.

Inputs: no named parameters.

genlayer_call_contractExecute GenLayer gen_callA

Execute a live gen_call request for read, write-simulation, or deploy-simulation against the configured GenLayer RPC endpoint.

Inputs: data (required), from (required), type (required), to, +4 more.

genlayer_eth_get_balanceGet EVM Balance Through GenLayerC
Read-onlyIdempotent

Call eth_getBalance through the configured GenLayer RPC endpoint.

Inputs: address (required), blockTag.

genlayer_explain_transaction_statusExplain GenLayer Transaction StatusA
Read-onlyIdempotent

Interpret a GenLayer transaction hash into appealability, finality phase, and likely next step.

Inputs: txId (required), timestamp.

genlayer_generate_agent_handoffGenerate Agent HandoffB
Read-onlyIdempotent

Generate an explicit step-by-step GenLayer handoff bundle for weaker agents, including ordered tool usage, fallback paths, and success conditions.

Inputs: code, goal, address, contractPath.

genlayer_generate_contract_playbookGenerate Contract PlaybookB
Read-onlyIdempotent

Generate a schema-aware deployment and interaction playbook for a GenLayer contract.

Inputs: code, address, contractPath.

genlayer_generate_typescript_workflowGenerate GenLayer TypeScript WorkflowA
Read-onlyIdempotent

Generate GenLayerJS deploy/read/write code from a contract schema or deployed contract.

Inputs: code, address, contractPath.

genlayer_get_contract_codeGet GenLayer Contract CodeA
Read-onlyIdempotent

Fetch the deployed base64 code blob for a GenLayer contract from the configured RPC endpoint.

Inputs: address (required), status, blockNumber.

genlayer_get_contract_interfaceGet GenLayer Contract InterfaceA
Read-onlyIdempotent

Normalize a contract schema into constructor, view methods, and write methods.

Inputs: code, address.

genlayer_get_contract_schemaGet GenLayer Contract SchemaA
Read-onlyIdempotent

Derive the schema/interface for base64-encoded GenLayer contract code.

Inputs: code (required).

genlayer_get_contract_snapshotGet GenLayer Contract SnapshotA
Read-onlyIdempotent

Fetch contract state, deployed code, and derived schema in one call.

Inputs: address (required), status, blockNumber.

genlayer_get_contract_stateGet GenLayer Contract StateA
Read-onlyIdempotent

Fetch the live state blob for a deployed GenLayer contract from the configured RPC endpoint.

Inputs: address (required), status, blockNumber.

genlayer_get_doc_by_slugGet GenLayer Doc By SlugA
Read-onlyIdempotent

Read a GenLayer documentation section by exact slug, path, resource URI, or docs URL.

Inputs: slug (required), maxChars.

genlayer_get_transaction_receiptGet GenLayer Transaction ReceiptA
Read-onlyIdempotent

Fetch the full receipt for a processed GenLayer transaction.

Inputs: txId (required).

genlayer_get_transaction_statusGet GenLayer Transaction StatusA
Read-onlyIdempotent

Fetch the lightweight consensus status for a GenLayer transaction.

Inputs: txId (required), timestamp.

genlayer_get_workflow_sessionGet Workflow SessionA
Read-onlyIdempotent

Read a persisted GenLayer workflow session by id.

Inputs: sessionId (required).

genlayer_inspect_transactionInspect TransactionB
Read-onlyIdempotent

Combine GenLayer status, GenLayer receipt, and Ethereum transaction lookup for a transaction hash.

Inputs: txId (required), timestamp.

genlayer_lint_contractLint Intelligent ContractA
Read-onlyIdempotent

Static pre-deploy checks for a GenLayer contract. Catches the mistakes that make a deploy finalize with a bare invalid_contract or fail consensus: a comment directly under the runner header, unpinned/alias runners, missing gl.Contract, and GenVM Python-subset issues (for/sorted/lambda) and storage anti-patterns.

Inputs: code (required).

genlayer_list_networksList GenLayer NetworksA
Read-onlyIdempotent

List current GenLayer network presets and RPC endpoints from the documented network matrix.

Inputs: no named parameters.

genlayer_list_sectionsList GenLayer Doc SectionsA
Read-onlyIdempotent

List available GenLayer documentation sections, optionally filtered by prefix text.

Inputs: limit, prefix.

genlayer_list_topicsList GenLayer TopicsA
Read-onlyIdempotent

List top-level GenLayer documentation topics with section counts and example pages.

Inputs: limit.

genlayer_list_workflow_sessionsList Workflow SessionsA
Read-onlyIdempotent

List persisted GenLayer workflow sessions ordered by most recently updated.

Inputs: limit.

genlayer_load_contract_artifactLoad Contract ArtifactA
Read-onlyIdempotent

Load a local contract artifact or bytecode file and return base64 plus file metadata for downstream GenLayer workflows.

Inputs: contractPath (required).

genlayer_metricsGet GenLayer MetricsA
Read-onlyIdempotent

Fetch Prometheus-style metrics from the configured GenLayer HTTP /metrics endpoint.

Inputs: no named parameters.

genlayer_network_statusGet GenLayer Network StatusA
Read-onlyIdempotent

Fetch a combined snapshot of GenLayer RPC connectivity, chain id, block height, and syncing status.

Inputs: no named parameters.

genlayer_node_healthCheck GenLayer Node HealthA
Read-onlyIdempotent

Fetch the configured GenLayer node HTTP health endpoint and return the live response.

Inputs: no named parameters.

genlayer_plan_contract_actionPlan Contract ActionA
Read-onlyIdempotent

Build a schema-validated execution plan for deploy, read, or write contract actions.

Inputs: action (required), args, code, from, +5 more.

genlayer_plan_contract_workflowPlan Contract WorkflowA
Read-onlyIdempotent

Build a multi-phase GenLayer contract workflow covering deploy, wait, snapshot, first read/write, and diagnosis.

Inputs: code, address, contractPath.

genlayer_probe_endpoint_capabilitiesProbe Endpoint CapabilitiesA
Read-onlyIdempotent

Probe the configured GenLayer deployment and report which health, RPC, sync, and debug surfaces are actually exposed.

Inputs: no named parameters.

genlayer_raw_rpcCall GenLayer RPC MethodC
Idempotent

Call a GenLayer JSON-RPC method directly against the configured endpoint.

Inputs: method (required), params.

genlayer_read_docRead GenLayer DocB
Read-onlyIdempotent

Read a specific GenLayer documentation section by slug, path, title, or fuzzy query.

Inputs: section (required), maxChars.

genlayer_refresh_docsRefresh GenLayer DocsA
Idempotent

Force-refresh the cached GenLayer documentation bundle from the configured source.

Inputs: no named parameters.

genlayer_run_contract_reportRun Contract ReportB
Read-onlyIdempotent

Orchestrate network context, contract snapshot, interface summary, workflow plan, and default action plans into one contract report.

Inputs: address (required).

genlayer_run_transaction_reportRun Transaction ReportA
Read-onlyIdempotent

Orchestrate waiting, inspection, status explanation, optional trace lookup, and optional contract snapshot into one transaction report.

Inputs: txId (required), wait, address, timestamp, +3 more.

genlayer_scaffold_contractScaffold Intelligent ContractA
Read-onlyIdempotent

Generate a working starter GenLayer intelligent contract for a chosen template. The output has the runner header pinned and avoids the common deploy-killers.

Inputs: template (required), name.

genlayer_scaffold_testScaffold Direct-Mode TestA
Read-onlyIdempotent

Generate a fast in-memory direct-mode test (genlayer-test) for a contract template. Uses the real fixtures (direct_vm, direct_deploy, direct_alice) and mocks web/LLM calls.

Inputs: template (required), name.

genlayer_search_docsSearch GenLayer DocsB
Read-onlyIdempotent

Search the GenLayer documentation bundle and return the most relevant sections.

Inputs: query (required), limit.

genlayer_search_examplesSearch GenLayer ExamplesA
Read-onlyIdempotent

Search GenLayer documentation sections that contain code blocks, commands, SDK usage, or configuration examples.

Inputs: query (required), limit.

genlayer_start_workflow_sessionStart Workflow SessionB

Create a persisted GenLayer workflow session from a generated contract workflow plan.

Inputs: code, goal, notes, address, +1 more.

genlayer_submit_raw_transactionSubmit Raw TransactionA

Submit a signed GenLayer or Ethereum-compatible raw transaction through eth_sendRawTransaction.

Inputs: rawTransaction (required).

genlayer_syncingCheck GenLayer Sync StatusB
Read-onlyIdempotent

Call the configured GenLayer RPC endpoint gen_syncing method.

Inputs: no named parameters.

genlayer_trace_transactionTrace GenLayer TransactionA
Read-onlyIdempotent

Call gen_dbg_traceTransaction for a transaction hash when the target node exposes debug methods.

Inputs: txID (required), round.

genlayer_update_workflow_stepUpdate Workflow StepA
Idempotent

Mark a specific workflow session step as completed or pending.

Inputs: sessionId (required), stepIndex (required), phaseIndex (required), completed.

genlayer_wait_for_transactionWait For TransactionA

Poll GenLayer transaction status until accepted or finalized, then return the combined inspection payload.

Inputs: txId (required), timestamp, intervalMs, maxAttempts, +1 more.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A3.6/5.0

    Scored across 46 tools

    Disambiguation4/5

    Most tools have distinct purposes, but there is some overlap among transaction-related tools (e.g., inspect, status, receipt, explain) and contract state retrieval tools (state, code, snapshot, schema). This could cause an agent to select the wrong tool without careful reading.

    Naming Consistency5/5

    All 46 tools follow a consistent 'genlayer_verb_noun' pattern with snake_case. The verb and noun are always clearly separated, making the tool set predictable and easy to navigate.

    Tool Count2/5

    With 46 tools, the surface is very large. While the scope covers many aspects of GenLayer development, the count exceeds the recommended 25+ threshold for 'too many', potentially overwhelming agents and increasing selection errors.

    Completeness5/5

    The tool set thoroughly covers the GenLayer lifecycle: scaffolding, linting, testing, workflow planning, contract deployment/interaction, network diagnostics, documentation retrieval, and transaction analysis. No obvious gaps are present.

    Related MCP Connectors

    Related MCP Servers

    • F
      license
      B
      quality
      D
      maintenance
      MCP server for AI agents to work with GenLayer projects, providing tools to run GenLayer CLI, lint contracts, and deploy contracts with automatic private key handling.
      4
      7 npm
      1
      -
    • A
      license
      A
      quality
      B
      maintenance
      A local-first MCP server that gives AI coding agents persistent memory and controlled commands. Features a git-backed markdown knowledge vault with FTS5 search, surgical section edits, token-aware context budgeting, and a sandboxed command engine with human approval gates. Works with Claude Code, Cursor, Copilot, Gemini, and more.
      62
      601 npm
      1
      Apache 2.0
    • A
      license
      A
      quality
      C
      maintenance
      Local-first MCP server that gives any AI coding agent per-project memory, workflow intelligence, and always-on, lossless token & context optimization.
      37
      31 npm
      6
      MIT