specwiki-checkpoint

A rule for completing code-changing tasks with a fixed quality checkpoint called HARNESS §0.3. The checkpoint requires test, security, review, quality-gate, risk, and manual-validation details.

In plain words
What is it for?
Use it when finishing code work or when explicitly asked to run the quality gate; it does not apply to ordinary questions or documentation-only changes.
Why use it?
It prevents a task from being declared complete without recording verification steps and known gaps.

Cursor rule for Cursor

Install

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.

agentmods
npx agentmods add rules/lucasviola/specwiki/specwiki-checkpoint
Clone the repo
git clone --depth 1 https://github.com/lucasviola/specwiki

Made for: Cursor.

Per session 549 This file is loaded in full into every session.
When invoked 549 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00549 $0.00549
Opus 5 $0.00275 $0.00275
Sonnet 5 $0.00110 $0.00110
Haiku 4.5 $0.00055 $0.00055

Measured 2d ago against content hash de77fe0d8da0, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

specwiki-checkpoint 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 2d 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.

.cursor/rules/specwiki-checkpoint.mdc · 47 lines

How it starts

The opening of the file, as written. The whole thing — 47 lines — stays where its author put it; the contents beside it link to each section on GitHub.

specwiki — Story completion checkpoint (HARNESS §0.3)

When finishing a code-changing story or task, or when the owner explicitly requests a quality-gate run, end the user-facing message with the full HARNESS §0.3 checkpoint — not a shortened summary.

Do not run the quality gate, automated code review, QA analysis, or return a §0.3 checkpoint for non-code prompts or non-executable metadata changes. This includes questions, discovery, and updates such as Markdown documentation or sprint-status YAML. Respond with a concise result instead.

Required sections (in order)

  1. What was built — 2–4 bullets
  2. Tests written — list new/changed test cases
  3. Logging (§0.8) — checklist with [x] or [ ]
  4. Security (§0.9) — checklist with [x] or [ ]
  5. Quality gate — all six commands with pass/fail and counts
  6. Automated code review (§0.2.5) — reviewer model + findings table (or "skipped" with reason)
  7. QA analysis (§0.2.6) — AC coverage, regression risks, gaps
  8. Manual validation stepsMANDATORY; never omit
  9. IMPLEMENTATION.md updated — yes/no
  10. Implement review patches? — ask owner before commit

Manual validation steps — non-negotiable

After the quality gate section, always include a ### Manual validation steps block with numbered, copy-pasteable commands and expected outcomes.

Sources (use in order):

  1. Story file ## QA Manual Validation section (if present)
  2. Story **Demo path:** in Dev Notes
  3. Generate steps from HARNESS §0.2.6 template tailored to the task

Each step must be: N. \command` — expected outcome`

Do not bury manual steps only in the story file. The owner must see them in the chat response.

Common specwiki commands

npm test
npm test -- tests/output/wiki.test.ts -t "pattern"
npm run dev generate -- --project tests/fixtures/sample-project --output /tmp/specwiki-qa
npm run dev generate -- --project tests/fixtures/sample-project --verbose 2>&1 | head
npm run dev open -- --project tests/fixtures/sample-project --output /tmp/specwiki-qa

Read the full file on GitHub · 47 lines

Changes

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.

  1. 2d ago First seen · 47 lines · 549 tokens per session scan A de77fe0d8da0

Subscribe to this mod's changes

specwiki-checkpoint is a cursor rule published in the GitHub repository lucasviola/specwiki (11 stars, last pushed 1mo ago), licensed MIT. It adds 549 tokens to every session, about $0.0027 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 2026-08-30.