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/st0012/cctop/agents-mdgit clone --depth 1 https://github.com/st0012/cctopWhat 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 | $0.02107 | $0.02107 |
| Opus 5 | $0.01053 | $0.01053 |
| Sonnet 5 | $0.00421 | $0.00421 |
| Haiku 4.5 | $0.00211 | $0.00211 |
Grade A, and why
cctop 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 3d 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.
How it starts
The opening of the file, as written. The whole thing — 139 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md - Development Guide for cctop
Scope and Sources
cctop is a macOS menu bar app that monitors AI coding sessions and jumps to them.
Use this file for rules that apply to most cctop work. Use these sources for detailed decisions:
- Read PRODUCT.md for product intent, feature fit, UX principles, and user-facing language.
- Read DESIGN.md for visual design rules.
- Read docs/session-files.md for session identity, visibility, and focus-routing contracts.
- Read docs/session-lifecycle.md for lifecycle, persistence, archive, and cleanup rules.
- Read site/README.md for website deployment and synchronization rules.
- Read video/framework.md for the video build, quality verification, and publication workflow.
Keep canonical repository guidance in this file. Pointer files, such as CLAUDE.md, must point to this file instead of copying it.
Required Development Rules
- Do not change the user's tool configuration without explicit consent. This includes plugins, hooks, and settings files.
- Do not make a compatibility break that requires running sessions to restart before they can reconnect.
- New features can remain unavailable to an existing session until that session restarts.
- Treat
~/.cctop/sessions/*.json, cctop logs, and client state as debugging evidence. A UI screenshot is only a symptom. - Keep app liveness, client liveness, visibility, archive state, hidden state, subagent state, and hook provenance separate.
- If a shared source changes, update all owners in the same change. Shared sources include schemas, hook versions, install paths, and release tooling.
- For every version change, run
scripts/bump-version.sh <version>. Do not edit version values separately. - Preserve canonical identity and order in
SessionManager.userSessions. DeriveSessionDataonly at the final presentation or direct-action boundary. A direct row action must use the exactSessionDatavalue that the row shows. - Keep Stream Deck downstream of the app. It reads
~/.cctop/display-state.jsonand must not recreate session policy.
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.
- 3d ago First seen · 139 lines · 2,107 tokens per session scan A dd60e92109da
cctop AGENTS.md is an instructions file published in the GitHub repository st0012/cctop (147 stars, last pushed 15d ago), licensed MIT. It adds 2,107 tokens to every session, about $0.0105 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 2026-08-30.
Other instructions, from other repositories
thinkrail AGENTS.md
Instructions for JetBrains/thinkrail, covering thinkrail, module structure & boundaries (top-priority requirement), engine: pi only, in-process, architecture (three rings) and repo layout.
PokeTokenBar CLAUDE.md
Instructions for chattymin/PokeTokenBar, covering poketokenbar — claude 프로젝트 지침, 참조 문서 (필요할 때 읽는다), 기여 언어 규약 (오픈소스 대비 — english first), 릴리스 (자연어 트리거) and 확장 규약 (새 프로바이더/툴 추가 시).
metis AGENTS.md
Instructions for Wholiver/metis, covering agent development guide, objective, start every task, repository map and desktop frontend changes (mandatory).
here-macos CLAUDE.md
Instructions for koalaauto/here-macos, covering here — ai assistant notes, architecture at a glance, prerequisites, build & run and always pass -quiet from an agent session.
signalbox CLAUDE.md
Instructions for dwmkerr/signalbox, covering signalbox, specs are the source of truth - keep them current, layout, build & test and regenerating the hero gif.
Plonk AGENTS.md
Instructions for ostapondo/Plonk, covering agent rules, layout, adding a module, build & verify and code style.