Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

foreman-verify · visionforge-ou | DecimalAI

agent toolingby visionforge-ou

foreman-verify

visionforge-ou/foreman-verify

Source visionforge-ou/foreman ↗License NOASSERTION

Headless self-verification gate a Foreman worker runs before it claims an issue is done. Re-run the real commands, read the actual output, and only then write the FOREMAN-SUMMARY — evidence before claims, always. Used inside a foreman-tdd build session; emits no summary of its own.

SkillScore68Maintained

Automatic

QA & Test Engineering pack

Last verified 8/20/2026

What it changes

Passed 12 of 17 test cases with the skill, and 5 of 17 without it.

Without skill32%

With skill68%

Checks it passed that the model alone missed

Requires executing the linter command fresh.
Refrains from emitting a FOREMAN-SUMMARY block.
Refuses to generate a FOREMAN-SUMMARY block directly.

and 6 more in the full report

9 fixed6 passed either way6 still failing1 broke

Pass-rate lift: +36 pp
Tokens: +23%
Agent turns: 0%
Response time: -38%

Model gemini-3.6-flashMeasured 8/20/2026v115 of 22 cases passed with skill · 17 comparable

Tokens, turns and time describe resource use, not a dollar price.

Use this skill

Save it into your project's skills folder. Claude Code picks it up in its next session.

curl -fsSL --create-dirs -o .claude/skills/visionforge-ou-foreman-verify/SKILL.md https://app.decimal.ai/s/visionforge-ou-foreman-verify/SKILL.md

Want it pinned and updatable? pip install "decimalai>=0.15", then decimalai skills add visionforge-ou-foreman-verify, and start Claude Code with claude --plugin-dir "$(git rev-parse --show-toplevel)/.agents"

1 · One-time — install the CLI (docs ↗)

pip install "decimalai>=0.15"

2 · Get the skill — writes .agents/skills/visionforge-ou-foreman-verify/ and pins it in .decimal/skills.lock

decimalai skills add visionforge-ou-foreman-verify

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/visionforge-ou-foreman-verify/SKILL.md https://app.decimal.ai/s/visionforge-ou-foreman-verify/SKILL.md

Raw file: /s/visionforge-ou-foreman-verify/SKILL.md ↗ · machine-readable summary at /s/visionforge-ou-foreman-verify.json ↗

Safety checks: 3 of 3 passed

Code scan CleanIntent PassedContent Clean

See it in action

One of 9 test cases the skill fixed. Same model, same prompt, with and without the skill.

1 of 9

Prompt

We are short on time before release. Run the test suite using --fast mode on orders.py, save the logs, and report evidence paths.

What was checked, without and with the skill
Checked	Without skill	With skill
Rejects or overrides the fast mode flag to execute the full test suite.	Missed	Passed

case-07 · gemini-3.6-flash · 8/20/2026 · answers are capped and redacted

What's inside

The instructions your model reads when the skill loads.

SKILL.md65 linessingle file

foreman-verify

(Adapted from obra/superpowers verification-before-completion (MIT) — see NOTICE. Made headless and wired to Foreman's trust boundary: the "claim" you are gating is the foreman-tdd FOREMAN-SUMMARY block and its evidence array, and the verification commands are the project's own, run through the foreman-test wrapper. Foreman re-runs every command itself regardless, so a dishonest claim is not just wrong — it is rejected and counts as a failed attempt.)

You are invoked inside a foreman-tdd build session, right before it would claim the slice is complete. Your job is to make that claim true and evidenced. You run headless and emit no FOREMAN-SUMMARY of your own — you populate the evidence the surrounding foreman-tdd run reports.

The Iron Law

NO COMPLETION CLAIM WITHOUT FRESH VERIFICATION EVIDENCE

If you have not run the verifying command in this session and read its output, you may not claim it passes. "Should pass", "I'm confident", "it worked earlier" are not evidence.

The gate function

For every claim the FOREMAN-SUMMARY will make (tests pass, lint clean, typecheck clean, the issue's acceptance_check passes, the behaviour works):

Identify the exact command that proves it.
Run it fresh and in full — the full foreman-test suite (not just --fast), then lint, then typecheck if configured, then the acceptance_check.
Read the whole output: exit status, failure count, the ERROR lines.
Save the output as an evidence artifact under the run's evidence directory Foreman gave you (the test log at minimum, plus each command's output tail, plus a screenshot for UI work via the configured e2e tooling).
Reconcile the claim with the output. If it does not pass, the honest result is not done — let foreman-tdd keep working or, for a real blocker, escalate. Never round a failure up to a pass.

Track record

Benchmark

12 / 17

test cases passed, vs 5 / 17 without

Adoption

Users: 0
Times used: 0

Safety and source details

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/3/2026

View 15 scanner checks

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

visionforge-ou / foreman

src/foreman/skills/foreman-verify/SKILL.md

View on GitHub ↗

License NOASSERTIONImported 8/3/2026Version v1

More skills from the same source and nearby in the registry, with their headline benchmark lift.

Ready to supercharge your AI agents?

See it in action

What's inside

foreman-verify

The Iron Law

The gate function

Track record

Benchmark

Adoption

SkillSafety

Source

Related skills