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/yonatangross/orchestkitWrote 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/rules/yonatangross/orchestkit/done-signoff)<a href="https://agentmods.dev/rules/yonatangross/orchestkit/done-signoff"><img src="https://agentmods.dev/badge/rules/yonatangross/orchestkit/done-signoff/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/rules/yonatangross/orchestkit/done-signoff"><img src="https://agentmods.dev/badge/rules/yonatangross/orchestkit/done-signoff.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 | $0.00024 | $0.00674 |
| Opus 5.5 | $0.00010 | $0.00270 |
| Sonnet 5.5 | $0.00005 | $0.00135 |
| Haiku 4.5 | $0.00002 | $0.00067 |
Grade A, and why
done-signoff 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 8d 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 — 69 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Done is a sign-off, not a claim
Use after verification-gate.md has produced fresh evidence. The gate proves the
work; the sign-off lets the person accept it. Skills that load this rule: verify
(always, at the end of its report) and glyph --signoff (appended to any render).
The gate
Lead with any FAIL. Then ask one question with exactly these three options, in this order, with these labels:
AskUserQuestion(questions=[{
"question": "<work>: <grade line>. Accept as done?",
"header": "Sign-off",
"multiSelect": False,
"options": [
{"label": "Accept done", "description": "Verified; close it out."},
{"label": "Show me the evidence", "description": "Re-run one check live, then ask again."},
{"label": "Not satisfied", "description": "Something is off; say what to re-check."}
]
}])
<grade line> is the verdict in the caller's own scale. From verify, that is the composite
and the verdict, for example 7.8/10, IMPROVEMENTS RECOMMENDED (scores are 0 to 10). From
glyph --signoff or any skill without a grade, it is the count of checks that ran, for example
11 of 12 checks passed.
- "Accept done": the work is closed. Not before.
- "Show me the evidence": re-run the named check in the main thread, paste the output, then ask again.
- "Not satisfied": treat the answer as a bug report. Fix, re-verify, ask again.
Incorrect
All 12 checks pass. Done!
The agent closed its own task. Nobody accepted it, and a check that failed earlier in the run is not in the reply.
Correct
❌ 1 of 12 failed: docs build (exit 1, generate-search-index).
✅ 11 of 12 passed (tests 412/412, typecheck 0 errors).
Then the gate above, with <grade line> = 11 of 12 checks passed (or, from verify,
6.4/10, BLOCKED).
When to skip the gate
- Non-interactive runs (
claude -p, aWorkflowscript, a background subagent) cannot callAskUserQuestion. Return the verdict as data ({passed, total, findings[]}) and let the main thread ask. ORK_ASK_FALLBACK=textis set: ask the same three options as a numbered text list with the same labels.- Pure conversation or a one-line doc edit: no gate.
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.
- 8d ago First seen · 69 lines · 24 tokens per session scan A 1104d20ea8e8
done-signoff is a cursor rule published in the GitHub repository yonatangross/orchestkit (295 stars, last pushed today), licensed MIT. It adds 24 tokens to every session and 674 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 2026-10-08.
Other cursor rules, from other repositories
unit-test-manners
When implementing unit tests, adhere to the following principles to ensure robust, maintainable, and fast tests.
ponytail
Ponytail, lazy senior dev mode. Always pick the simplest solution that works.
angular-20
This rule provides comprehensive best practices and coding standards for Angular development, focusing on modern TypeScript, standalone components, signals, and performance optimizations.
dev-standard
Apache Superset development standards and guidelines for Cursor IDE.
prefer-assertions-over-defensive-checks
Prefer assertions over defensive checks when data is guaranteed to be valid.
prefer-direct-imports-over-module-mocks
Prefer extracting a testable core over vi.mock / vi.resetModules when unit tests need to reach production logic entangled with config, env, or singletons.