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/andreaswasita/copilot-agents-dojonpx agentmods add skills/andreaswasita/copilot-agents-dojo/verify-before-doneWrote 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/andreaswasita/copilot-agents-dojo/verify-before-done)<a href="https://agentmods.dev/skills/andreaswasita/copilot-agents-dojo/verify-before-done"><img src="https://agentmods.dev/badge/skills/andreaswasita/copilot-agents-dojo/verify-before-done/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/andreaswasita/copilot-agents-dojo/verify-before-done"><img src="https://agentmods.dev/badge/skills/andreaswasita/copilot-agents-dojo/verify-before-done.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00020 | $0.01741 |
| Opus 5.5 | $0.00008 | $0.00696 |
| Sonnet 5.5 | $0.00004 | $0.00348 |
| Haiku 4.5 | $0.00002 | $0.00174 |
Grade A, and why
verify-before-done 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 1mo 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 — 165 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Verify Before Done Skill
Refuses to mark a task complete without concrete evidence from both automated checks and the real artifact or user-facing surface. Produces an explicit VERIFIED, NOT VERIFIED, or INCONCLUSIVE verdict. Does NOT replace scripts/verify.sh — it tells the agent when to run it and what evidence the gate cannot supply.
When to Use
- Before flipping any step in
tasks/todo.mdfrom- [ ]to- [x]. - Before opening a pull request.
- After any fix or feature implementation.
- After performance work, migrations, or changes whose success depends on runtime behavior.
- Any time the agent is about to say "done" or "ready" without evidence.
Prerequisites
- A test runner installed for the stack (pytest, vitest, go test, etc.).
scripts/verify.sh(orscripts/run-checks.ps1on Windows) executable.- The
powershelltool to invoke runners andgit. gitavailable so diffs can be inspected.
How to Run
1. State the success predicate and capture a baseline when behavior is changing.
2. Run the project's tests and the dojo gate via the `powershell` tool.
3. Exercise the real artifact or user-facing surface.
4. Inspect the diff and working tree.
5. Record evidence and assign a verdict.
6. Mark the step complete only when the verdict is `VERIFIED`.
Quick Reference
| Check | Command (run via powershell) |
Pass criterion |
|---|---|---|
| Tests pass | pytest / npm test / go test ./... / dotnet test / mvn test |
exit 0, no failures reported |
| Dojo gate | bash scripts/verify.sh --check |
exit 0 |
| Real artifact | Matching browser, CLI, API, package, migration, or runtime check | observed result satisfies the stated predicate |
| Baseline comparison | Same measurement before and after the change | result moves from failing/old state to expected state |
| Diff summary | git diff main --stat |
matches plan in tasks/todo.md |
| No regressions | git diff main -- <touched paths> |
only intended changes |
| Clean tree | git status --porcelain |
empty output (or expected untracked) |
| Evidence captured | edit tasks/todo.md |
Verification Results block added |
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.
- 1mo ago Changed · +45 lines e620722899fe
- 1mo ago First seen · 120 lines · 20 tokens per session scan A 530692c5b92d
verify-before-done is a skill published in the GitHub repository andreaswasita/copilot-agents-dojo (54 stars, last pushed 1mo ago), licensed MIT. It adds 20 tokens to every session and 1,741 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 2026-08-30.
Other skills, from other repositories
Verification & Quality Assurance
Comprehensive truth scoring, code quality verification, and automatic rollback system with 0.95 accuracy threshold for ensuring high-quality agent outputs and codebase reliability.
Verification & Quality Assurance
Comprehensive truth scoring, code quality verification, and automatic rollback system with 0.95 accuracy threshold for ensuring high-quality agent outputs and codebase reliability.
gate
MANDATORY before every commit and push. All gates must pass — Gates 1–3 always run; Gates 4 (Docker) and 5 (Cloud) run only when applicable. Do NOT proceed if any gate fails — fix the issue first.
Verification Before Completion
Mandatory empirical verification suite execution before declaring task completion.
verification-quality
Use when you need to verify work quality using truth scoring (0.0–1.0) with environment-tiered thresholds and machine-readable JSON CI gate output.
Verification & Quality Assurance
Comprehensive truth scoring, code quality verification, and automatic rollback system with 0.95 accuracy threshold for ensuring high-quality agent outputs and codebase reliability.