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/longbkit/clisbot/claude-mdgit clone --depth 1 https://github.com/longbkit/clisbotWhat 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 | $0.01889 | $0.01889 |
| Opus 5 | $0.00945 | $0.00945 |
| Sonnet 5 | $0.00378 | $0.00378 |
| Haiku 4.5 | $0.00189 | $0.00189 |
Grade A, and why
clisbot CLAUDE.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.
How it starts
The opening of the file, as written. The whole thing — 191 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Scope
These rules apply to everything inside this repository.
The stable implementation contract lives in:
docs/architecture/architecture-overview.mddocs/architecture/surface-architecture.mddocs/architecture/runtime-architecture.mddocs/architecture/domain-language.md
If implementation conflicts with those docs:
- stop
- refactor toward the docs if the fix is clear
- ask the user before proceeding if the conflict changes behavior, architecture, or scope
Do not silently drift away from the architecture docs.
First Read Order
Load context in this order unless the task is obviously narrower:
README.mddocs/overview/README.md- the architecture docs listed above
Then load only the smallest extra context that matches the task:
- feature work:
docs/features/README.md, the relevant feature doc, then linked task docs - task execution:
docs/tasks/README.md,docs/tasks/backlog.md, then the relevant task doc - operator or onboarding or release work:
docs/development/README.mdand the relevant user-guide doc - research-heavy questions: the relevant file under
docs/research/
Do not front-load the whole docs tree.
Documentation Precedence
When documents disagree, use this order:
docs/architecture/README.md,docs/development/README.md, anddocs/user-guide/docs/features/docs/tasks/docs/research/anddocs/lessons/
docs/research/ and docs/lessons/ are supporting context only. They should not silently override architecture or guide docs.
Repo Map
Use this ownership map before editing:
src/channels: Slack and Telegram surfaces, route handling, rendering, pairing, follow-up behaviorsrc/agents: durable agent state, sessions, queueing, loops, attachments, run lifecyclesrc/auth: roles, permissions, owner claim, authorization resolutionsrc/config: schema, loading, credentials, templatessrc/control: operator CLI, runtime lifecycle, health, status, logs, bootstrapsrc/runners: execution backends, currently tmuxsrc/infra: small host/runtime primitives only, with no product conceptstest/: regression and behavior coveragedocs/tests/: readable validation scenarios when behavior needs explicit ground truth
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 · 191 lines · 1,889 tokens per session scan A 6e3421aacbb0
clisbot CLAUDE.md is an instructions file published in the GitHub repository longbkit/clisbot (114 stars, last pushed 2d ago), licensed MIT. It adds 1,889 tokens to every session, about $0.0094 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
HARTOS CLAUDE.md
Instructions for hertz-ai/HARTOS, covering claude.md, project overview, master key - ai exclusion zone, branch discipline — main branch only (mandatory) and sibling repos — canonical filesystem paths.
sample-ai-assistant-on-agentcore AGENTS.md
AGENTS.md instructions for aws-samples/sample-ai-assistant-on-agentcore, covering agents.md — sparky ai assistant on bedrock agentcore, project structure, prerequisites, build and run commands and frontend.
Clouisle AGENTS.md
AGENTS.md instructions for clouisle/Clouisle, covering agents.md, project overview, top-level directories, core commands and backend.
Clouisle CLAUDE.md
Claude Code instructions for clouisle/Clouisle: Read and follow AGENTS.md before working in this repository.
intelligent-terminal rust.instructions.md
Concise Rust coding conventions for this repository.
Browser4 CLAUDE.md
Instructions for platonai/Browser4, covering browser4 — project context for claude, architecture, key dispatch chain (cli → browser), batch commands and e2e test structure.