Borrowing it
Nothing to install: this file belongs to ctxpipe-ai/ctxpipe. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/ctxpipe-ai/ctxpipe/main/AGENTS.mdgit clone --depth 1 https://github.com/ctxpipe-ai/ctxpipeWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/ctxpipe-ai/ctxpipe/agents-md)<a href="https://agentmods.dev/instructions/ctxpipe-ai/ctxpipe/agents-md"><img src="https://agentmods.dev/badge/instructions/ctxpipe-ai/ctxpipe/agents-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/ctxpipe-ai/ctxpipe/agents-md"><img src="https://agentmods.dev/badge/instructions/ctxpipe-ai/ctxpipe/agents-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.08405 | $0.08405 |
| Opus 5.5 | $0.03362 | $0.03362 |
| Sonnet 5.5 | $0.01681 | $0.01681 |
| Haiku 4.5 | $0.00840 | $0.00840 |
Grade C, and why
ctxpipe AGENTS.md scanned grade C with 3 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 13d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Asks for rootmediumPrivilege escalation
A mod that escalates privileges can change anything on the machine, not only the project.
- **Docker image**: the environment is built from [`.agents/Dockerfile`](.agents/Dockerfile) following Cursor’s **Running Docker** guidance ([Cloud Agent setup](https://cursor.com/docs/cloud-agent/setup)): Docker CE + `f Reads MCP configurationmediumAgent snooping
mcp.json carries server URLs and auth tokens; reading it lets a mod discover and abuse other integrations.
- **MCP:** [`.mcp.json`](.mcp.json) mirrors `.cursor/mcp.json` in Claude Code syntax (`${VAR:-}`), with API-key auth for headless use: `CTXPIPE_API_KEY` (sent as `x-api-key`; empty falls back to OAuth), `NEON_API_KEY`, ` Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- **Sandbox quirks:** no IPv6, so the hook writes a gitignored `docker-compose.override.yml` that turns off FalkorDB's Bolt listener, which the backend never uses. Container egress is TLS-intercepted, so Docker image bui How it starts
The opening of the file, as written. The whole thing — 253 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Where agent instructions live
Agent instructions are distributed: this file covers repo-wide rules; apps and packages can have their own AGENTS.md with local instructions. When working in an app or package, read both the root AGENTS.md and that folder's AGENTS.md (if present).
- Root (this file): architecture, code style, and Simplified Technical English for explanations.
- apps/backend: apps/backend/AGENTS.md — API, OpenAPI, MCP, Drizzle, TypeScript, etc. Source-connectors skill when designing, building, or reviewing a source connector (Linear, Notion, Slack, git-native mirror/capture,
connections.config, self-host), when provider requests can multiply per related record, or when a full import must resume after a crash. - apps/otel-collector: laptop OTLP debug sink for
pnpm dev:infra(stdout only). Hosted ingest isops/observability(ADR-038). - apps/codesearch: apps/codesearch/AGENTS.md — Zoekt/SCIP orchestration, read-only DB, OpenAPI + Zod, and the manual Kubernetes ingest memory gate.
- apps/ui: apps/ui/AGENTS.md — TanStack Start frontend, React Aria, Tailwind, Storybook, Vitest; React skill when building or editing components; product-ui skill and DESIGN.md when building or restyling product screens.
- apps/docs: apps/docs/AGENTS.md — Fumadocs documentation site (Next.js 15, Shiki, forced-dark, deploys to docs.ctxpipe.ai).
- examples/: runnable consumer examples for ctxpipe packages (manual e2e tests against real infra). See examples/README.md; first entry is examples/aws-cdk-self-host for
@ctxpipe-ai/aws-cdkon AWS. - ops/observability: internal ClickStack + Langfuse Railway project (not product deploy). See ops/observability/README.md and ADR-038.
- plugins/ctxpipe: Claude plugin for the hosted product MCP. See ADR-026.
MCP (project-scoped): Config lives at .cursor/mcp.json (same file as .agents/mcp.json via the .agents → .cursor symlink). Two kinds of servers:
- Product MCP (
ctxpipe) — hosted org MCP for the product itself (https://app.ctxpipe.ai/mcp?orgSlug=ctx-tev). This is the customer-facing ctxpipe tool surface agents use against the live org; do not point it at localhost. - Agent tooling MCPs — editor/agent helpers for this repo (Storybook, Neon, Railway, HyperDX, Langfuse). Not the backend’s in-process Hono
/mcpproduct server.
| Server | Purpose | Preconditions |
|---|---|---|
ctxpipe |
Hosted product MCP (org ctx-tev) |
OAuth / account access to that org in Cursor |
ctxpipe-storybook |
Storybook MCP at http://127.0.0.1:6006/mcp |
Storybook must be running: pnpm --filter @ctxpipe/ui storybook |
neon |
Neon Lakebase Postgres via hosted MCP — read-only (?readonly=true) |
OAuth in Cursor (uncheck Full access if prompted; URL param forces RO). Optional headless: Bearer NEON_API_KEY + same readonly=true URL (Neon MCP docs) |
railway |
Railway status + deploy/runtime stdout | OAuth in Cursor (type: streamable-http → https://mcp.railway.com) |
hyperdx |
ClickStack logs, traces, metrics, dashboards | HYPERDX_ACCESS_KEY = HyperDX personal access key (Bearer). UI https://hyperdx.ctxpipe.ai. Ingest token HYPERDX_API_KEY is a different secret |
langfuse |
Langfuse project MCP (LLM traces / prompts) | LANGFUSE_AUTH_STRING = base64(pk:sk). URL is https://langfuse.ctxpipe.ai/api/public/mcp. See ops/observability/USING.md |
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.
- 13d ago Changed · +21 lines · +795 tokens per session scan B → C c543ae6f8d5a
- 17d ago Changed · +4 lines · +234 tokens per session 5b7a8ab06178
- 18d ago Changed · +20 tokens per session 18304bb1c17e
- 19d ago Changed · +13 tokens per session 9b2cf411fa24
- 20d ago Changed · +11 lines · +415 tokens per session c5fa1eff5220
- 22d ago Changed · +4 lines · +115 tokens per session a20ddc93fcb4
- 1mo ago Changed · +1 lines · +43 tokens per session 2e73fd453da7
- 1mo ago First seen · 212 lines · 6,770 tokens per session scan B c06701b5a3c4
ctxpipe AGENTS.md is an instructions file published in the GitHub repository ctxpipe-ai/ctxpipe (26 stars, last pushed today), licensed MIT. It adds 8,405 tokens to every session, about $0.0336 per session on Opus 5.5. A static security scan graded it C with 3 findings (asks for root, reads mcp configuration, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
vscode buildNext.instructions.md
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
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, cache eligibility and rollout and applying the notice (cutover).
langchain AGENTS.md
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.
deepseek-harness AGENTS.md
AGENTS.md instructions for deepseek-ai/deepseek-harness, covering agents.md, pre-stable apis and released session data, repository layout, commands and host sandbox failures.
vscode copilot-instructions.md
Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).