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 commands/moazbuilds/claudeclaw/helpgit clone --depth 1 https://github.com/moazbuilds/claudeclawWhat 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 | $0.00004 | $0.00684 |
| Opus 5 | $0.00002 | $0.00342 |
| Sonnet 5 | $0.00001 | $0.00137 |
| Haiku 4.5 | $0.00000 | $0.00068 |
Grade A, and why
help 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.
What it actually says
Display this help information to the user:
ClaudeClaw — daemon mode plus one-shot prompt/trigger runs.
Commands:
/heartbeat:start— Initialize config and start the daemon/heartbeat:stop— Stop the running daemon/heartbeat:clear— Back up the current session and restart fresh/heartbeat:status— Show daemon status, countdowns, and config/heartbeat:config— View or modify heartbeat settings (interval, prompt, telegram)/heartbeat:jobs— Create, list, edit, or delete cron jobs/heartbeat:logs— Show recent execution logs (accepts count or job name filter)/heartbeat:telegram— Show Telegram bot status and sessions (useclearto reset sessions)/heartbeat:help— Show this help message
Start command options (CLI):
bun run src/index.ts start— normal daemon modebun run src/index.ts start --prompt "text"— one-shot prompt, no daemon loopbun run src/index.ts start --trigger— start daemon and run startup trigger oncebun run src/index.ts start --prompt "text" --trigger— start daemon and run startup trigger with custom prompt- Add
--telegramwith--triggerto forward startup trigger output to configured Telegram users - Add
--web(optional--web-port 4632) to start a local dashboard with the daemon
Send command options (CLI):
bun run src/index.ts send "text"— send to active daemon sessionbun run src/index.ts send "text" --telegram— send and forward output to Telegram- If daemon is already running, use
send;startwill abort.
How it works:
- The daemon runs in the background checking your schedule every 60 seconds
- A heartbeat prompt runs at a fixed interval (default: every 15 minutes)
- Jobs are markdown files in
.claude/claudeclaw/jobs/with cron schedules (timezone-aware, evaluated in configuredtimezone) - The statusline shows a live countdown to the next run
Configuration:
.claude/claudeclaw/settings.json— Main config (heartbeat, telegram, security).claude/claudeclaw/settings.json— Main config (heartbeat, telegram, security, web).claude/claudeclaw/jobs/*.md— Cron jobs with schedule frontmatter and a prompt body
Job file format:
---
schedule: "0 9 * * *"
---
Your prompt here. Claude will run this at the scheduled time.
Schedule uses standard cron syntax: minute hour day-of-month month day-of-week
Note: Bun is required to run the daemon. It will be auto-installed on first /heartbeat:start if missing.
Telegram:
- Configure in
.claude/claudeclaw/settings.jsonundertelegram - Daemon mode can run Telegram polling in-process when token is configured
- Startup trigger
start --trigger --telegramand daemonsend --telegramcan forward responses
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 · 60 lines · 4 tokens per session scan A e5a262d6513a
help is a command published in the GitHub repository moazbuilds/claudeclaw (1,313 stars, last pushed 1mo ago), licensed MIT. It adds 4 tokens to every session and 684 once invoked, about $0.0000 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 commands, from other repositories
git
Git operations with intelligent commit messages and workflow optimization.
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
constitution
Create or update the project constitution from interactive or provided principle inputs.