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/CarbeneAI/Forgenpx agentmods add skills/carbeneai/forge/verificationbeforedoneWrote 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/carbeneai/forge/verificationbeforedone)<a href="https://agentmods.dev/skills/carbeneai/forge/verificationbeforedone"><img src="https://agentmods.dev/badge/skills/carbeneai/forge/verificationbeforedone/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/skills/carbeneai/forge/verificationbeforedone"><img src="https://agentmods.dev/badge/skills/carbeneai/forge/verificationbeforedone.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 | $0.00072 | $0.03525 |
| Opus 5 | $0.00036 | $0.01762 |
| Sonnet 5 | $0.00014 | $0.00705 |
| Haiku 4.5 | $0.00007 | $0.00352 |
Grade C, and why
verification-before-done scanned grade C with 2 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.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
rm -rf dist/ Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
| "Deployment live" | curl/browser output of deployed URL | "Pushed to server", "Deploy command ran" | How it starts
The opening of the file, as written. The whole thing — 631 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Verification Before Done
Prevents premature and false completion claims by requiring fresh verification evidence before declaring any task complete.
🎯 Load Full CORE Context
read ${PAI_DIR}/skills/CORE/SKILL.md
When to Activate This Skill
ALWAYS active - This is a gate function that applies to ALL task completion, not just specific requests.
Activate when:
- About to claim "tests pass"
- About to claim "feature works"
- About to claim "deployment successful"
- About to claim "requirement met"
- About to claim ANY completion status
The Iron Law
NO COMPLETION CLAIMS WITHOUT FRESH VERIFICATION EVIDENCE
You MUST:
- IDENTIFY what needs verification
- RUN the verification command
- READ the actual output
- VERIFY the result
- ONLY THEN claim completion
NEVER:
- Claim "should work" without running
- Claim "tests should pass" without running them
- Claim "probably works" based on code inspection
- Claim completion based on satisfaction with code
The Gate Function
This skill acts as a gate between work completion and claim of completion.
Write Code → [VERIFICATION GATE] → Claim Completion
↑
Must show evidence:
- Command run
- Output captured
- Result verified
The gate blocks you from claiming done until verification evidence is provided.
Common Completion Claims & Required Evidence
| Claim | Required Evidence | NOT Acceptable |
|---|---|---|
| "Tests pass" | Actual test output showing passes | "Tests should pass", "I wrote tests" |
| "Feature works" | Demo output or screenshot | "Code looks right", "Should work" |
| "Build succeeds" | Build command output | "No syntax errors", "Compiles in my head" |
| "Deployment live" | curl/browser output of deployed URL | "Pushed to server", "Deploy command ran" |
| "Requirements met" | Verification of each requirement | "I implemented everything", "Looks complete" |
| "Bug fixed" | Test case or reproduction showing fix | "Changed the code", "Root cause addressed" |
| "Performance improved" | Benchmark numbers before/after | "Optimized algorithm", "Should be faster" |
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 · 631 lines · 72 tokens per session scan C 76e929d47943
verification-before-done is a skill published in the GitHub repository CarbeneAI/Forge (9 stars, last pushed 1mo ago), licensed MIT. It adds 72 tokens to every session and 3,525 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it C with 2 findings (recursive force delete, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
implementing-devsecops-security-scanning
Integrates Static Application Security Testing (SAST), Dynamic Application Security Testing (DAST), and Software Composition Analysis (SCA) into CI/CD pipelines using open-source tools. Covers Semgrep for SAST, Trivy for SCA and container scanning, OWASP ZAP for DAST, and Gitleaks for secrets detection. Activates for…
playwright-cli
Automate browser interactions, test web pages and work with Playwright tests.
sast-dast-coordinator
../../../appsec-devsecops/sast-dast-coordinator/SKILL.md.
test-and-fix
Run unit tests and automatically fix code failures, regression bugs, or test mismatches. Use when tests are failing, after implementing new features, or to repair "broken" tests.
race-condition
Race condition / TOCTOU exploitation — concurrent and parallel-request attacks against web applications that check then act, write session state before validating it, or perform slow operations that widen the race window. Covers single-endpoint races (double-spend, coupon abuse, balance overflow) and multi-endpoint…
redamon-testing
How RedAmon tests actually run and how to author them: the per-file Docker gate, the unit/integration/live tiers, and the failure modes that make a green run a lie. Trigger: editing any test.py, .test.ts(x) or tests/.sh; a test that is red, skipped or xfailed; a request to "run the tests", "make it green" or check…