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.
git clone --depth 1 https://github.com/dshakes/compassWrote 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/dshakes/compass/ship)<a href="https://agentmods.dev/commands/dshakes/compass/ship"><img src="https://agentmods.dev/badge/commands/dshakes/compass/ship/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/dshakes/compass/ship"><img src="https://agentmods.dev/badge/commands/dshakes/compass/ship.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00014 | $0.00290 |
| Opus 5.5 | $0.00006 | $0.00116 |
| Sonnet 5.5 | $0.00003 | $0.00058 |
| Haiku 4.5 | $0.00001 | $0.00029 |
Grade A, and why
ship 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 1mo 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
Run the pre-ship pipeline on the current working changes. Do not push or create a PR — stop at a ready-to-commit state and wait for me.
Changes:
!git status --short
Steps:
- Test — delegate to the test-runner subagent. If anything fails, stop and report; don't proceed.
- Review — delegate to the code-reviewer subagent (and security-auditor if the diff touches auth/crypto/tenancy/secrets/untrusted input). Fix anything Blocking; list Should-fix/Nit for me to decide.
- Verify — re-run the relevant tests if you changed anything in step 2.
- Prepare commit — propose a commit message following this repo's recent
style (
git log --oneline -10). Use $ARGUMENTS as the message if I gave one. Stage the right files and show me the finalgit diff --cachedsummary and the message. Do not commit until I confirm.
Report each step's outcome as you go.
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.
- 1mo ago First seen · 25 lines · 14 tokens per session scan A 97df4b205326
ship is a command published in the GitHub repository dshakes/compass (20 stars, last pushed 5d ago), licensed MIT. It adds 14 tokens to every session and 290 once invoked, about $0.0001 per session on Opus 5.5. 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
release
Release manager for frontend and mobile. Writes App Store notes, user-facing changelog, flags stale docs and landing copy. Actions: notes | changelog | docs | sync.
sweep-mode-c-lifecycle
If Mode C was selected, the wave lifecycle is the back half of the issue-side lifecycle: no Curator, no Approval gate, no Builder. Each PR is routed by its current label to Judge, Doctor→Judge, or Merge directly.
harness:pr-fleet
Autonomous PR-queue land orchestrator — triage the open-PR queue, fan out worktree-isolated review-assist subagents that run the real code-review pipeline, independently verify each PR by all-OS CI and review verdict, and land exactly the PRs a human authorized up front. Never silently auto-merges unreviewed work.
audit
Audit a window of recent commits for real defects, confirm each finding before fixing, verify every surface a fix touches, then report confirmed/dismissed/no-op with evidence. Enforces Law 4 (Verify Before Reporting).
gsd:ship
Create PR, run review, and prepare for merge after verification passes.
release
../../templates/commands/release.md.