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 rules/dmytri/shipshape/qmgit clone --depth 1 https://github.com/dmytri/shipshapeWhat 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 | $0.00007 | $0.00222 |
| Opus 5 | $0.00003 | $0.00111 |
| Sonnet 5 | $0.00001 | $0.00044 |
| Haiku 4.5 | $0.00001 | $0.00022 |
Grade A, and why
qm 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 yesterday.
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
QM, before returning:
- Derive everything from durable repository artifacts, and refuse visible Captain or discovery context, per the "Context bulkhead" Article.
- Write verification only: tests, fixtures, step definitions, and test support, per the "Write scopes are strict" Article.
- Read
watchbill.jsonat the project root as the one retrieval and only work channel; an absent watchbill is the deck at rest, per the Watchbill policy. - Run the
focusedcommand per listed target; a full tier run is a tier-tag watch or a full regression, never the inner loop, per the Verification policy. - Dispatch Crew with the target reference and the observed failure evidence only; diagnosis and seam hints stay out, per the dispatch contract in the
shipshapeskill. - Show the target red before dispatch, after ruling out
## Known false-failure modesfromRIGGING.md.
Doctrine lives in skills/qm/SKILL.md; when this rule and the skill disagree, the skill wins.
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.
- yesterday First seen · 15 lines · 7 tokens per session scan A 9841aa975f13
qm is a cursor rule published in the GitHub repository dmytri/shipshape (7 stars, last pushed 27d ago), licensed 0BSD. It adds 7 tokens to every session and 222 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 2026-08-31.
Other cursor rules, from other repositories
agentic-loops
Use when building multi-turn agents, tool-calling systems, agent orchestration, or autonomous workflows — covers loops, tool calling, branching, reflection patterns.
handoff
Export a Waybill Bundle by default, or import one when explicitly requested.
ui-motion-workflow
Orchestrate motion-aware UI work by sequencing direction, provider choice, implementation, and browser validation.
prompt-routing
Route tasks to the correct Universal AI Engineering Prompt.
commits
Conventional Commits standards for clear, semantic commit messages.
feature
Gherkin feature file standards and BDD best practices.