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.
npx agentmods add instructions/limjonathan/runtime-mcp/agents-mdgit clone --depth 1 https://github.com/limjonathan/runtime-mcpWhat 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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00701 | $0.00701 |
| Opus 5 | $0.00351 | $0.00351 |
| Sonnet 5 | $0.00140 | $0.00140 |
| Haiku 4.5 | $0.00070 | $0.00070 |
Grade A, and why
runtime-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 yesterday.
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.
How it starts
The opening of the file, as written. The whole thing — 59 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md — runtime-mcp
Project
Drop-in MCP library exposing a running Node app's live state (logs, DB schema, config, feature flags) to coding agents. Monorepo of 4 packages.
Layout
packages/
├── core/ # runtime-mcp: registry, endpoint, discovery, redaction, SQL guard
├── connect/ # runtime-mcp-connect: stdio shim agents register in .mcp.json
├── integrations/ # adapters: express/pino/winston/pg/prisma/node-sqlite (subpath exports)
└── demo-app/ # Express + node:sqlite demo with planted bugs
Session-start workflow
Before deep work, regenerate AI indexes if stale:
npx repomix . --output repomix-output.xml
graphify update . --no-cluster
Use repomix-output.xml and graphify query instead of reading files one-by-one.
Commands
pnpm install # after changing deps
pnpm -r build # tsup builds all packages (core must build before connect/integrations typecheck)
npx vitest run # full suite from repo root (39 tests)
cd packages/demo-app && npx tsc --noEmit # demo has noEmit tsconfig
Per-package typecheck: cd packages/<name> && npx tsc --noEmit.
Conventions
- TypeScript strict +
noUncheckedIndexedAccess+verbatimModuleSyntax— useimport type, watch for possibly-undefined index access. - ESM everywhere; relative imports need
.jsextensions in TS source. - No comments in code; no console.log in library code (console hook is the product).
- Immutable patterns; no mutation of shared state objects.
- SDK is MCP TypeScript SDK v2:
McpServer.registerTool(name, config, cb),registerResource(name, uriOrTemplate, config, cb), templates vianew ResourceTemplate(uri, { list })(list key REQUIRED, may be undefined). Transports:NodeStreamableHTTPServerTransport(enableJsonResponse: truefor deterministic responses), client via@modelcontextprotocol/client(Client,StreamableHTTPClientTransport, authProvider{ token: async () => token }). - Per-session hosting pattern: create transport →
server.connect(transport)→transport.handleRequest(req,res,body); session map keyed bymcp-session-id.
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.
- yesterday First seen · 59 lines · 701 tokens per session scan A 5cf809b54f71
runtime-mcp AGENTS.md is an instructions file published in the GitHub repository limjonathan/runtime-mcp (0 stars, last pushed 10d ago), licensed MIT. It adds 701 tokens to every session, about $0.0035 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.
Other instructions, from other repositories
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
buildNext
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
next.js AGENTS.md
Instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
spec-kit AGENTS.md
Instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.
langchain AGENTS.md
Instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.