What is ctxlayer AGENTS.md?

A developer guide for CtxLayer, a web application built with a Cloudflare Worker, a service that runs code at the network edge, plus a React web interface. It gives the standard change, testing, and file-location rules.

Use it when changing worker routes, database migrations, session or collaboration storage, the web interface, shared data formats, or project tests.

It helps contributors find the right part of the application and verify changes locally before sending them for review.

ctxlayer AGENTS.md — Instructions file by stevenn

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/stevenn/ctxlayer/agents-md

Clone the repo

git clone --depth 1 https://github.com/stevenn/ctxlayer

Made for: Codex, OpenCode.

Per session 1,825 This file is loaded in full into every session.

When invoked 1,825 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.

Model	Per session	Once invoked
Fable 5	$0.01825	$0.01825
Opus 5	$0.00912	$0.00912
Sonnet 5	$0.00365	$0.00365
Haiku 4.5	$0.00183	$0.00183

Measured 2d ago against content hash db0c316a010a, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

ctxlayer 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 · 141 lines

How it starts

The opening of the file, as written. The whole thing — 141 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Working in this repo as an agent

Optimised for Claude-on-the-web and mobile sessions. Keep replies terse, operate via slash commands and bun run scripts, and prefer one-shot verifies over interactive flows.

Golden path for any change

bun run verify — typecheck + lint (Biome) + unit + integration tests, all offline. bun run verify:full additionally runs the smoke suite, which needs a running local Worker (bun run dev:worker) or a preview URL.
Commit with a present-tense, "why" focused message.
git push -u origin <branch> and open a PR with gh pr create.
There is no hosted CI — run bun run verify yourself before pushing.

Where things live

Cloudflare Worker entry → apps/worker/src/index.ts
Durable Objects → apps/worker/src/mcp/session-do.ts, apps/worker/src/collab/doc-room-do.ts
Hono REST routes → apps/worker/src/api/*.ts
SQL migrations → apps/worker/src/db/migrations/*.sql
React SPA → apps/web/src/
Shared types/schemas → packages/shared/src/
OKF (Open Knowledge Format) interop → packages/shared/src/frontmatter.ts
- apps/worker/src/docs/okf.ts; the doc rail is the frontmatter editor. Reference → docs/plan/M-okf.md
Architecture reference → docs/PLAN.md (reference, not a roadmap)

Module rules

One file = one concern. Split when a file holds separable concerns — not when it crosses a line count.
~300 LoC is a review trigger, not a limit. Past it, ask "is this still one thing?" A standalone component, a single DO's lifecycle, or one linear pipeline stays one file however long it runs; db/queries/docs.ts (three tables + ACL predicates) and mcp/tools-proxy.ts (registry + catalogue read-models + call runner) are the shape that should split.
No circular deps across the four top-level concerns: api/, mcp/, collab/, queues/.
All env via the typed Env from apps/worker/src/env.ts.
All SQL via apps/worker/src/db/queries/*.ts — never inline in routes.

Read the full file on GitHub · 141 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.

2d ago First seen · 141 lines · 1,825 tokens per session scan A db0c316a010a

Subscribe to this mod's changes

ctxlayer AGENTS.md is an instructions file published in the GitHub repository stevenn/ctxlayer (9 stars, last pushed 4d ago), licensed MIT. It adds 1,825 tokens to every session, about $0.0091 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.

Frontend Security Testing ai-agents cloudflare-workers durable-objects hono mcp model-context-protocol

ctxlayer AGENTS.md

Getting it into your agent

What it costs to keep this loaded

Grade A, and why

Nothing flagged

How it starts

Working in this repo as an agent

Golden path for any change

Where things live

Module rules

What this file has done since we first saw it

Other instructions, from other repositories

projektor AGENTS.md

projektor CLAUDE.md

capacities-mcp-worker CLAUDE.md

vibesdk AGENTS.md

nutrition-mcp CLAUDE.md

munkel CLAUDE.md