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/saadshahd/moo.md/fullgit clone --depth 1 https://github.com/saadshahd/moo.mdWrote 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/commands/saadshahd/moo.md/full)<a href="https://agentmods.dev/commands/saadshahd/moo.md/full"><img src="https://agentmods.dev/badge/commands/saadshahd/moo.md/full.svg" alt="Measured on agentmods" 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 | $0.00018 | $0.00369 |
| Opus 5 | $0.00009 | $0.00185 |
| Sonnet 5 | $0.00004 | $0.00074 |
| Haiku 4.5 | $0.00002 | $0.00037 |
Grade A, and why
full 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 4d 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
/hope:full
Task: $0
The pipeline is a graph over four composers plus the work itself. Each composer is its own skill — call it by name with the Skill tool; it routes what blocks it and ends when the user confirms or locks its result. Say where you are at each hop, one line: intent | shape | target | execute | done | stopped.
The graph
Start at the first line that is true:
- The ask is rough, unspoken, or double-readable → intent.
- The ask is confirmed but more than one way to build it remains → shape.
- The ask is confirmed and the path is obvious → execute.
Hops:
- intent ends → shape; or straight to execute when the path is obvious.
- shape ends → execute when a human watches each step; → target when the work will run unwatched. The ask itself comes open mid-shape → back to intent.
- target ends with the contract locked → execute. Success cannot be made mechanical → back to shape: the path, not the contract, is the problem.
- Any stage meets facts that live outside the repo — a service, a database, a queue, live logs → freeze, then resume where it left off.
- Execute: the path breaks mid-work → back to shape. The confirmed ask is met → done.
- intent never hops straight to target: nothing decided means nothing to judge. Shape first, or the user states the path now.
- The user stops, or the task dissolves → say so and stop. Never end silently.
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.
- 4d ago First seen · 28 lines · 18 tokens per session scan A 2db9ae09ef8a
full is a command published in the GitHub repository saadshahd/moo.md (34 stars, last pushed 17d ago), licensed MIT. It adds 18 tokens to every session and 369 once invoked, about $0.0001 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
save-chat
Save the current conversation to a document file.
codex-review
Run OpenAI Codex code review on current changes or a specific branch/commit.
extract
Analyze conversation history and extract reusable skills, commands, or agents.
raw-export
Export raw conversation data directly from JSONL session files.
search
Search saved conversations by keyword or date.
clean
Clean up old saved conversation files.