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 skills add dayvisonassis/sdd-skills --skill evaluatorgit clone --depth 1 https://github.com/dayvisonassis/sdd-skillsWrote 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/dayvisonassis/sdd-skills/evaluator)<a href="https://agentmods.dev/skills/dayvisonassis/sdd-skills/evaluator"><img src="https://agentmods.dev/badge/skills/dayvisonassis/sdd-skills/evaluator/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/dayvisonassis/sdd-skills/evaluator"><img src="https://agentmods.dev/badge/skills/dayvisonassis/sdd-skills/evaluator.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 | $0.00117 | $0.09588 |
| Opus 5.5 | $0.00047 | $0.03835 |
| Sonnet 5.5 | $0.00023 | $0.01918 |
| Haiku 4.5 | $0.00012 | $0.00959 |
Grade A, and why
evaluator 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 16d 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 — 336 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Evaluator
A second, independent layer of validation. After implement-feature finishes a feature, the evaluator looks at it from the outside and confirms — against contract.md — whether the delivery conforms. It finds deviations the implementer's self-check missed. It does not fix code; it evaluates and orchestrates the correction loop, dispatching fix-runner on failure.
Base docs:
https://github.com/dayvisonassis/sdd-skills/blob/main/docs/Skill_Evaluator.md(full rationale),https://github.com/dayvisonassis/sdd-skills/blob/main/docs/Contrato_de_Feature.md(contract structure),https://github.com/dayvisonassis/sdd-skills/blob/main/docs/Como_criar_gates.md(gates),https://github.com/dayvisonassis/sdd-skills/blob/main/docs/ARQUITETURA_DAS_SKILLS.md(flow, states, progress.json schema). Report schema:references/evaluation-report-schema.md.
The evaluator complements automated tests — it does not replace them.
INPUT
Free-form. The skill needs:
- The target feature (ID/name) — explicit and required. Abort if absent or ambiguous (list candidates).
- Auto-discovers:
progress.json(root ofdocs/) and the feature'scontract.md(indocs/<feature-id>-<kebab>/). - Optional free-form overrides at the end — e.g. "no screenshots", "gates only", "max 5 attempts" (overrides
maxFixAttemptsfor this run), "no exception: " (a human refused the gate exceptionfix-runnerasked for; see the schema, "Gate exceptions").
If the feature has no contract.md, abort: "No contract.md for F — generate it with spec-writer first."
OUTPUT
- Screenshots — visual evidence of observable UI criteria (when applicable).
- Report — consolidated ✓ / ✗ / — per contract criterion and gate.
- Findings in chat — textual summary for the user.
- State in
progress.json— CLEAN | FAIL | PENDING | ABORTED for the feature, committed when the evaluation ends (Step 9, which has the one exception). - On FAIL: a structured
evaluation-report.jsonin the feature folder, with each failure classified bykind(consumed byfix-runnerfor code, or a test-writer for tests). Removed when the feature ends CLEAN. - Coverage-table dispute entries — the only thing the evaluator writes in the writer's checklist: it opens a
disputedentry when it disagrees with the writer, closes one that became moot and removes one that went stale (Disputes lifecycle,../e2e-test-writer/references/e2e-rules.md). - Test commits — the test files a validator confirmed, with their checklists, in commits of their own, and with them a code correction a commit hook held back (Step 8.4). Writers never commit, so this is what keeps the committed tree passing its own gates.
- The evaluator does not edit code or tests itself. Code corrections →
fix-runner; test corrections → the matching test-writer (confirmed by the matching test-validator).
What ships with it
1 file 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.
- 16d ago Changed · +65 lines c8f956b2d069
- 18d ago Changed · +3 lines b33bf0032b8d
- 25d ago Changed 09a28f22f82d
- 29d ago Changed · +21 lines · +117 tokens per session 649b74dac95c
- 1mo ago Changed · +43 lines c4e0732466ae
- 2mo ago First seen · 204 lines · 0 tokens per session scan A a739f3d9b5ad
evaluator is a skill published in the GitHub repository dayvisonassis/sdd-skills (3 stars, last pushed 16d ago), licensed MIT. It adds 117 tokens to every session and 9,588 once invoked, about $0.0005 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 2026-08-31.
Other skills, from other repositories
test-master
Generates test files, creates mocking strategies, analyzes code coverage, designs test architectures, and produces test plans and defect reports across functional, performance, and security testing disciplines. Use when writing unit tests, integration tests, or E2E tests; creating test strategies or automation…
critique
Adversarial design critique of the current work. Render it, look at it, and argue for rejection. Run after the gates are green, never instead of them.
ship
Pre-release gate. Run the full gate, responsive + render checks, then produce the release checklist (README badge/current/changelog). Use before tagging a release.
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
ai-eval-plan
Design an evaluation plan for an LLM or AI feature before shipping it. Use when asked how to evaluate a prompt/model/agent, set up an eval harness, define quality metrics for an AI feature, or build a regression gate. Produces an eval plan — task definition, datasets, metrics & rubrics, baselines, automated + human…
browser-test-loop
Use to actually verify a web change in a real browser instead of assuming it works — click through a flow, catch console errors, check a form submits, confirm a fix on the page. Trigger when a UI change needs proof, a bug only reproduces in the browser, or the user asks to test, check, or screenshot the running app.