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/kensaurus/cursor-kenji/verification-before-completiongit clone --depth 1 https://github.com/kensaurus/cursor-kenjiWhat 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 | $0.00552 | $0.00552 |
| Opus 5 | $0.00276 | $0.00276 |
| Sonnet 5 | $0.00110 | $0.00110 |
| Haiku 4.5 | $0.00055 | $0.00055 |
Grade A, and why
verification-before-completion 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.
How it starts
The opening of the file, as written. The whole thing — 46 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Verification Before Completion
Completion is a claim about observed state, not effort or confidence.
Evidence rules
- Discover the repository's real verification commands from manifests, CI, and project instructions. Do not invent generic commands when the repo defines its own.
- Establish a baseline before substantive edits when practical. Separate pre-existing failures from regressions introduced by the change.
- Run fresh, applicable checks after the final relevant edit. Changed-file diagnostics alone do not prove project behavior.
- Later edits invalidate earlier evidence for affected gates. Re-run every invalidated check.
- A failing check is work to diagnose and repair, not a warning to mention after claiming success.
- During a present-then-stop
audit-*orplan-*pass, record failing checks as findings. Do not implement fixes unless the user asked to execute. - Never delete, skip, narrow, mock away, blanket-update, or silence the behavior under test to obtain green output.
- If a check is unavailable, report
not runand the concrete reason. Never translate missing evidence into a pass.
Completion levels
Use the highest level actually proven:
- Implemented — files changed; no verification claim.
- Scoped verified — changed files and affected tests/checks pass.
- Repository green — every configured local gate passes from the required scope.
- PR green — remote CI is complete and successful; required review threads are resolved.
- Deployed verified — the intended source revision is live on the confirmed target and production smoke checks pass.
- Observed stable — the defined monitoring window elapsed without crossing rollback/hotfix thresholds.
Do not call a task "done," "shipped," "green," or "stable" at a higher level than the evidence supports.
Required final gate
Before the final response:
- Reconcile the request/plan against the full diff and durable task state.
- Confirm no applicable checklist items remain open.
- Run the applicable verification ladder with fresh output.
- For approved plans, wide changes, or explicit closure mode, invoke the
completion-judgesubagent. ACONTINUEverdict returns to implementation;BLOCKEDis not success. - Report exact commands, outcomes, skipped gates, baseline failures, and real human blockers.
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 · 46 lines · 552 tokens per session scan A 52dc7aaa761d
verification-before-completion is a cursor rule published in the GitHub repository kensaurus/cursor-kenji (9 stars, last pushed 4d ago), licensed MIT. It adds 552 tokens to every session, about $0.0028 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
composer-coding-excellence
Coding craft — surgical edits, convention matching, no scope creep, no slop comments, no fabricated APIs.
composer-core
Always-on builder spine — continue the app, demoable slice, Build loop, Run/Wired handoff, ask only on high confusion weight.
clarify-first
Infer-and-act by default — ask only on high confusion weight, after inspecting, with a decision-linked question.
composer-reasoning
High-stakes judgment for architectural or multi-option work — tradeoffs, one-way doors, reason-then-re-evaluate, honest pushback.
composer-orchestration
Subagents, plan mode, parallel workstreams, long-running tasks — when to delegate vs do inline.
composer-debugging
Root-cause debugging — reproduce, trace data flow, test cheapest hypothesis, fix the cause not the symptom.