Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/momentmaker/kaijutsunpx agentmods add skills/momentmaker/kaijutsu/verification-before-completionWrote 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/skills/momentmaker/kaijutsu/verification-before-completion)<a href="https://agentmods.dev/skills/momentmaker/kaijutsu/verification-before-completion"><img src="https://agentmods.dev/badge/skills/momentmaker/kaijutsu/verification-before-completion.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 5.1 | $0.00067 | $0.01273 |
| Opus 5 | $0.00034 | $0.00636 |
| Sonnet 5 | $0.00013 | $0.00255 |
| Haiku 4.5 | $0.00007 | $0.00127 |
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 7d 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 — 119 lines — stays where its author put it; the contents beside it link to each section on GitHub.
verification-before-completion
Adapted from
obra/superpowers/skills/verification-before-completionunder the MIT License. Copyright (c) Jesse Vincent. Modifications by kaijutsu maintainers — pared down for kaijutsu's "trust the swarm" stance and reframed as a primitive other skills compose.
A discipline gate. Stops you from claiming work is done based on vibes. Forces evidence.
This skill is composed by other skills via deps.skills. Also directly invocable when you catch yourself about to say "should pass".
What this is NOT
- Not a finding-triage gate. v0.7 quality fingerprinting + multi-agent swarm consensus already handles that.
- Not human ceremony layered on agent output. The agents are trusted; what we're verifying is YOUR CODE, not the agents' findings.
- Not a process for everything. A README typo fix doesn't need a verification gate. Use judgment — see "When to skip" below.
What this IS
A short checklist between "I think this is done" and "I'm telling the user it's done". The cost of a stale claim is high: lost trust, wasted review cycles, debugging time tomorrow on bugs you said were fixed today.
When to invoke
- Before saying "tests pass" / "linter clean" / "build succeeds" / "bug fixed"
- Before committing
- Before opening / merging a PR
- After auto-fixers (eslint --fix, prettier, etc.) — tools fail silently more than you think
- When
polishorincremental-implementationsays "clean" — convergence ≠ correctness - After ANY agent (subagent, swarm, etc.) reports success — verify independently
When to skip
- Pure prose / comment / doc edits with no executable impact
- Research / explore / read-only work
- The verification command itself is what's being changed (chicken-and-egg — note this and let user verify)
The gate
BEFORE claiming any status:
1. IDENTIFY: What command proves this claim?
2. RUN: Execute the FULL command, fresh, this turn (not "I ran it earlier")
3. READ: Full output. Check exit code. Count failures.
4. VERIFY: Does output confirm the claim?
NO → state actual status with evidence
YES → state claim WITH evidence
5. ONLY THEN: emit the claim
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 7d ago First seen · 119 lines · 67 tokens per session scan A 4c9752b870d5
verification-before-completion is a skill published in the GitHub repository momentmaker/kaijutsu (3 stars, last pushed 2mo ago), licensed MIT. It adds 67 tokens to every session and 1,273 once invoked, about $0.0003 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 skills, from other repositories
e2e
Generate and run Playwright E2E tests traced to spec.md acceptance criteria, with an optional accessibility audit. Use when saying "e2e tests" or "a11y audit".
tdd-cycle
Execute full TDD red-green-refactor cycle with validation gates. Use when saying "TDD cycle", "test-driven development", or "full TDD workflow".
done
Close an increment: ledger check, specweave verify, optional review, then specweave complete. Use when all tasks are done and saying "close increment", "we are done", or "finish up".
genie-orca-review
Independent, read-only review of a group, a wish, or a PR on Orca — SHIP / FIX-FIRST / BLOCKED with severity-tagged findings. Council and retro are this skill with a different input.
argot-check
Score your working changes with argot — flag code foreign to this repo's own patterns (unfamiliar dependencies, APIs, constructs), functions the repo already has, code filed in the wrong place, imports that break the repo's layering, and tests weakened, disabled, or deleted alongside a production change — before…
atomic-tdd
Test-first discipline. Auto-triggers on "let's implement X", "add feature Y", "fix bug Z", "write a test for", "implement", "build out", and similar pre-code-change phrases. Iron rule: failing test exists before production code. Skip only for pure docs/config changes with an explicit "skipped because:" note. Explicit…