# AGENTS.md — Build guide for Composio

## Project scope
Expose three hand-written MCP tools for the user's chosen APIs, with typed arguments, explicit credentials and a preview/approval step for writes.

Catalogue verdict: kinda. If you personally use four apps, not four hundred, most of Composio evaporates. An agent can write you a local MCP server that wraps Gmail, Slack, GitHub and Notion with your own personal access tokens in about a session, and the tool-calling loop works fine. What you are actually buying from Composio is the part nobody enjoys: OAuth app registration per provider, token refresh, per-user credential storage, rate limit handling, and schemas for tools you have not thought of yet. That gets ugly the moment you want more than a handful of integrations, or multiple users, or an OAuth flow instead of a static token you pasted in. Solo and token-based: build it. Multi-tenant or breadth-hungry: pay.
Use the implementation prompt below to define the deliverable. Complete each phase's acceptance checks before extending the scope.

## Working agreement
- Inspect the repository and its existing instructions before choosing paths, dependencies or commands. Keep one coherent stack and explain changes to the proposed architecture.
- Plan a vertical slice that accepts a real input and produces the useful output described below. Persist only the state the prompt calls for; respect memory-only and upstream-managed workflows. Use fixtures only when they are clearly labelled.
- After scaffolding, document the actual install, development, check and build commands in README and keep them synchronized with the package or project manifest. Do not report commands as successful unless they ran.
- Work in small steps. At handoff, list implemented flows, checks actually performed, remaining blockers, and any credentials or provider setup the owner must supply.
- Do not publish, spend money, contact customers, delete source data or run irreversible migrations without the project owner's authorization.

## Prerequisites
- Runtime and tools: TypeScript, Node, PostgreSQL and a database-backed job worker with a small React run inspector.
- Before starting: Credentials for only the chosen integrations, their current API documentation, example payloads and an always-on worker for schedules.

## Stack and architecture
- TypeScript, Node, PostgreSQL and a database-backed job worker with a small React run inspector
- Data design: Store ToolDefinitionVersion, Invocation, CredentialReference and ExternalReceipt; read and write tools are distinct, and approval covers the exact tool arguments and destination.
- Setup: Credentials for only the chosen integrations, their current API documentation, example payloads and an always-on worker for schedules

## Security and data integrity
- Persist trigger identity, step input version and external receipts before marking success. Retry bounded transient failures; unknown write outcomes require reconciliation. Do not promise exactly-once execution across external services.
- Use the current official MCP SDK contract and the selected APIs only. Remote transport/auth differs from local stdio; do not advertise fifty connectors or general OAuth management from three adapters.
- Keep secrets outside client bundles and exported projects; document what leaves the device and make retention/deletion controls visible.

## Agent implementation rules
- Project rule — domain: Store ToolDefinitionVersion, Invocation, CredentialReference and ExternalReceipt; read and write tools are distinct, and approval covers the exact tool arguments and destination.
- Project rule — scope and recovery: Use the current official MCP SDK contract and the selected APIs only. Remote transport/auth differs from local stdio; do not advertise fifty connectors or general OAuth management from three adapters.
- Project rule — acceptance: Change a destination after approving a draft action; require a fresh approval and do not execute the old grant against the new target.
- Project rule — delivery: document real setup commands and permissions; do not claim a build, accuracy level, performance result or security certification that has not been demonstrated.

## Optional agent skills and references
- Recommended skill: [mcp-builder](https://github.com/anthropics/skills/blob/main/skills/mcp-builder/SKILL.md) — implement this actual MCP tool server with typed contracts and documented transport/permission boundaries. Follow the maintainer's installation instructions and match its requirements to the chosen runtime.
- Recommended skill: [sharp-edges](https://github.com/trailofbits/skills/blob/main/plugins/sharp-edges/skills/sharp-edges/SKILL.md) — review configuration and API defaults against the app-specific invariants and recovery boundaries above; this is not a security certification. Follow the maintainer's installation instructions and match its requirements to the chosen runtime.

Read the linked SKILL.md and its dependencies before adding a skill. Select only the skills matching this project's runtime and task; their documentation does not supply API access, credentials or approval to perform external actions. Pin the reviewed revision where the tool supports it. Follow the chosen agent's documented project-level installation mechanism.

## Distribution ideas
These are optional planning notes. Obtain the owner's approval before publishing or contacting anyone.
- Demonstrate this working slice using synthetic or explicitly authorized non-sensitive examples: Expose three hand-written MCP tools for the user's chosen APIs, with typed arguments, explicit credentials and a preview/approval step for writes.
- Share a synthetic example export and the acceptance walkthrough; keep real customer, health, financial and source data private: Change a destination after approving a draft action; require a fresh approval and do not execute the old grant against the new target.
- State the limits before asking someone to replace their existing tool: Use the current official MCP SDK contract and the selected APIs only. Remote transport/auth differs from local stdio; do not advertise fifty connectors or general OAuth management from three adapters.

## Engineering roadmap
1. Phase 1 — Pin the working slice and create its example input: Expose three hand-written MCP tools for the user's chosen APIs, with typed arguments, explicit credentials and a preview/approval step for writes. Confirm setup: Credentials for only the chosen integrations, their current API documentation, example payloads and an always-on worker for schedules.
2. Phase 2 — Implement persistence and write-time invariants before decorating the UI: Store ToolDefinitionVersion, Invocation, CredentialReference and ExternalReceipt; read and write tools are distinct, and approval covers the exact tool arguments and destination.
3. Phase 3 — Connect the working view to real saved state. Persist trigger identity, step input version and external receipts before marking success. Retry bounded transient failures; unknown write outcomes require reconciliation. Do not promise exactly-once execution across external services.
4. Phase 4 — Expose the app-specific limits and recovery path in context: Use the current official MCP SDK contract and the selected APIs only. Remote transport/auth differs from local stdio; do not advertise fifty connectors or general OAuth management from three adapters.
5. Phase 5 — Walk through this concrete acceptance case and preserve its exported evidence: Change a destination after approving a draft action; require a fresh approval and do not execute the old grant against the new target. Finish the README and backup/restore instructions; report unfinished capabilities explicitly.

## Paid-product capabilities outside this build
- Hundreds of pre-built, pre-tested tool schemas you did not have to read API docs for
- Managed OAuth: app registration, consent screens, refresh tokens, revocation
- Per-end-user credential storage, which is the whole game if your agent serves anyone but you
- Someone else fixing your integration when a provider ships a breaking API change
- Usage observability, retries and rate limit smoothing across providers

## Implementation prompt
Build the following focused alternative to Composio. This is a deliberately limited personal or small-team substitute, not parity with the paid service.

WORKING SLICE
Expose three hand-written MCP tools for the user's chosen APIs, with typed arguments, explicit credentials and a preview/approval step for writes.

SETUP AND ARCHITECTURE
Use TypeScript, Node, PostgreSQL and a database-backed job worker with a small React run inspector. Prerequisites: Credentials for only the chosen integrations, their current API documentation, example payloads and an always-on worker for schedules. Before integrating anything, record actual versions and permissions, plus model files or provider limits only where used, in the README; make unavailable dependencies visible rather than simulating success.

DOMAIN MODEL AND INVARIANTS
Store ToolDefinitionVersion, Invocation, CredentialReference and ExternalReceipt; read and write tools are distinct, and approval covers the exact tool arguments and destination.

IMPLEMENTATION CONTRACT
Persist trigger identity, step input version and external receipts before marking success. Retry bounded transient failures; unknown write outcomes require reconciliation. Do not promise exactly-once execution across external services. Provide an input/setup view, the main work view, and a review/export view appropriate to this workflow. Preserve the last saved state if a job or save fails. Include empty, loading, permission-denied, partial and retryable-error states. Log identifiers and error categories without secret values or unnecessary private content.

APP-SPECIFIC BOUNDARY AND RECOVERY
Use the current official MCP SDK contract and the selected APIs only. Remote transport/auth differs from local stdio; do not advertise fifty connectors or general OAuth management from three adapters.

ACCEPTANCE SCENARIO
Change a destination after approving a draft action; require a fresh approval and do not execute the old grant against the new target. Also reopen the app after an interrupted operation, confirm the saved record/export remains inspectable, and document the recovery action. These are implementation acceptance requirements, not a claim that this guide has been tested.

DELIVERY
Deliver a runnable repository with migrations or project-format versioning, a non-sensitive example, environment/permission setup, the exact manual acceptance steps, and a backup/export-and-restore walkthrough. Implement the working slice before optional integrations; list any deferred paid-product capabilities honestly. Do not add capabilities outside the working slice just to resemble the original product.

PROJECT RULES FOR AGENTS.md
Keep the domain invariants above executable at the write boundary. Propose scope changes before adding providers or permissions. Never fabricate source evidence, publish results, identity matches or successful delivery. Preserve user originals and require an explicit confirmation for destructive changes or external publication.

## Completion evidence
Demonstrate the prompt's acceptance scenarios against the scoped workflow. Include setup from a clean checkout and failure recovery. Check persistence across restart and export/restore only for the state the prompt says to store; for memory-only tools, confirm that temporary content is discarded as specified. Record actual results and remaining limitations. A detailed plan alone does not establish a working replacement.