session-collab-mcp AGENTS.md

Repository instructions for a TypeScript Model Context Protocol server that coordinates work between multiple local coding sessions using a SQLite registry.

In plain words
What is it for?
Use them to find the session-claim logic, MCP routing, database code, hooks, migrations, or tests, and to run targeted or complete test suites.
Why use it?
They explain the server's structure, tool locations, database behavior, and test commands so agents can modify it safely.

Instructions file for CodexOpenCode

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add instructions/leaf76/session-collab-mcp/agents-md
Clone the repo
git clone --depth 1 https://github.com/leaf76/session-collab-mcp

Made for: Codex, OpenCode.

Per session 1,160 This file is loaded in full into every session.
When invoked 1,160 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.01160 $0.01160
Opus 5 $0.00580 $0.00580
Sonnet 5 $0.00232 $0.00232
Haiku 4.5 $0.00116 $0.00116

Measured 2d ago against content hash 66c1add64e23, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

session-collab-mcp AGENTS.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

AGENTS.md · 137 lines

How it starts

The opening of the file, as written. The whole thing — 137 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AGENTS.md - Session Collab MCP

Generated: 2026-01-03 | Branch: master

Overview

TypeScript MCP server. 10 tools (v2 action-based). Local SQLite WIP registry for same-machine multi-session claims — not a remote lock.

Structure

src/
├── cli.ts              # Entry: stdio JSON-RPC server
├── constants.ts        # Version + short server instructions
├── hooks/              # Claude Code PreToolUse claim guard
├── db/
│   ├── db-path.ts      # SESSION_COLLAB_DB + modern/legacy paths
│   ├── queries.ts      # SQL (still monolithic; split later)
│   ├── sqlite-adapter.ts
│   ├── types.ts        # All type definitions
│   └── __tests__/      # In-memory SQLite tests
├── mcp/
│   ├── protocol.ts     # JSON-RPC types
│   ├── schemas.ts      # Zod validation (all inputs)
│   ├── server.ts       # Tool routing by prefix
│   └── tools/          # 4 files, 10 MCP tools ← see tools/AGENTS.md
├── utils/
│   ├── crypto.ts       # generateId()
│   └── response.ts     # errorResponse(), successResponse()
├── migrations/         # 12 SQL files, versioned schema
└── plugin/             # Claude Code plugin (hooks, skills, commands)

Where to Look

Task Location
Add MCP tool src/mcp/tools/ + register in server.ts
Add schema src/mcp/schemas.ts (Zod)
Add DB query src/db/queries.ts
Add type src/db/types.ts
Error codes src/utils/response.ts → ERROR_CODES
Test pattern src/db/__tests__/test-helper.ts

Commands

npm run build        # tsup (bundles all except better-sqlite3)
npm run typecheck    # tsc --noEmit (strict, noUnusedLocals)
npm run lint         # eslint
npm run test         # vitest

# Single test
npx vitest run src/db/__tests__/queries.test.ts
npx vitest run -t "should create a new session"

Conventions

Imports (ESM, .js required)

import type { DatabaseAdapter } from '../db/sqlite-adapter.js';  // Types first
import { createClaim } from '../db/queries.js';                   // Values second

Read the full file on GitHub · 137 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 137 lines · 1,160 tokens per session scan A 66c1add64e23

Subscribe to this mod's changes

session-collab-mcp AGENTS.md is an instructions file published in the GitHub repository leaf76/session-collab-mcp (2 stars, last pushed 2d ago), licensed MIT. It adds 1,160 tokens to every session, about $0.0058 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.