AGENT SKILL

design-system-coverage

Grades the design-system contract for completeness and writes the coverage inventory every later design-test skill reads. Use when a design-test run starts; it always runs first and its tier gates the run.

License 许可
MIT
所属仓库
Eliyce/paqad-ai
最近更新
2026-11-13
Stars
★ 8
Forks
0

01 前言 Front Matter

name
design-system-coverage
metadata
{ "cacheable": "false", "model_tier": "reasoning", "paqad-meta": "skill.meta.yaml", "output_format": "json" }
description
Grades the design-system contract for completeness and writes the coverage inventory every later design-test skill reads. Use when a design-test run starts; it always runs first and its tier gates the run.

02 资源文件 9

03 SKILL.md 正文

What It Does

Grades docs/instructions/design-system/ for completeness, produces a coverage inventory, and emits the tier (missing | bare | adequate | strong) the rest of the workflow uses to decide whether to stop, warn, or proceed in strict mode. Runs as Step 0 (readiness gate) and as the first skill in Step 1: every downstream skill consumes this inventory.

Use This When

As the first skill in every design-test run. The tier it emits gates the entire workflow. Never skip it; never run other design-test skills without it.

Inputs

  • Read every file under docs/instructions/design-system/ (typically tokens.md, components.md, accessibility.md, motion.md, patterns.md, responsive.md).
  • Read references/contract-clauses.md before grading.

Procedure

The decisions in steps 1–3 are deterministic: drive them with the scripts in scripts/ rather than re-deriving the logic from prose each run.

  1. Run scripts/list-contract-files.sh to enumerate which contract files exist and which are empty/stub.
  2. For each file, run scripts/count-clauses.sh <file> to count contract clauses (the file emits <path>\t<count> and skips headings, blank lines, frontmatter delimiters, and lines inside fenced code blocks). Concatenate the rows.
  3. Pipe the concatenated counts into scripts/derive-tier.sh: it emits a single tier=<missing|bare|adequate|strong> line. This is the gate decision: don't second-guess it.
  4. Emit the coverage inventory per assets/output.template.json to .paqad/design-test/runs/<run_id>/artifacts/contract-summary.json.
  5. Validate with scripts/validate-contract.sh: enforces required fields, allowed tier vocabulary, and rejects empty clause arrays when the tier is adequate or strong.
  6. If tier is missing, raise a Decision Pause Contract packet asking the user whether to invoke the documentation-update workflow inline for design-system before resuming. If tier is bare, additionally run scripts/gap-report.sh and emit each row as a DT-DS-XXXX documentation-drift finding (the script's id sequence is stable across runs so retests preserve ids).

Output Contract

  • Match assets/output.template.json: { tier, files: [{ path, present, empty, clause_count }], clauses: { tokens: [...], components: [...], accessibility: [...], responsive: [...], motion: [...], patterns: [...] } }.
  • tier ∈ missing | bare | adequate | strong.
  • Output must pass scripts/validate-contract.sh (exit 0).

Escalate / Stop Conditions

  • Stop and prompt when tier is missing. Offer inline invocation of the documentation-update workflow for the design-system.
  • Warn when tier is bare. Run downstream skills in exploratory mode and tag findings confidence: low.
  • Do not auto-engage design_test.strict when tier is strong: strict mode is explicit opt-in only.

Resources

  • references/contract-clauses.md
  • scripts/list-contract-files.sh: enumerate contract files (present/empty/non-empty).
  • scripts/count-clauses.sh: count clauses in one contract file.
  • scripts/derive-tier.sh: derive the tier from per-file counts (pure function of counts).
  • scripts/gap-report.sh: enumerate contract gaps as DT-DS-XXXX rows with stable ids.
  • scripts/validate-contract.sh: schema-check the emitted coverage inventory.
  • assets/output.template.json
  • agents/openai.yaml

04 同仓库更多 Skills

Eliyce
Eliyce/paqad-ai

sample-skill

Sample bundled skill fixture

★ 8
Eliyce
Eliyce/paqad-ai

processing-invoices

Turns a supplier invoice into one ledger line. Use when a supplier invoice arrives and needs booking.

★ 8
Eliyce
Eliyce/paqad-ai

processing-invoices

Turns a supplier invoice into one ledger line. Use when a supplier invoice arrives and needs booking.

★ 8
Eliyce
Eliyce/paqad-ai

understand-idea

Chat and Claude Code: Step 1 of the discover journey: clarifies the goal, who benefits, success and constraints into the idea block. Use when the journey directs it to step 1, not directly.

★ 8
Eliyce
Eliyce/paqad-ai

size-work

Chat and Claude Code: Suggest a size for a ticket from comparable finished work, as a labelled suggestion the team still owns. Use it when someone asks "how big is this?", "roughly how much work is ABC-12?", or wants a backlog item sized. It finds the team's sizing scale at first use, picks two to five of the closest finished tickets with links, and gives a size or range with the drivers behind it and a confidence of low, medium or high, lowered and explained when the area's documents are template text. Every result is labelled a suggested size, never a date or a commitment. It writes nothing into the size field unless the user says "set it", and it refuses to overwrite an existing size unless the user asks to replace it. Produces the sizing block of the record.

★ 8
Eliyce
Eliyce/paqad-ai

refine-ticket

Chat and Claude Code: Improve one or more existing tracker tickets without losing the author's intent. Use it when someone wants a ticket read, diagnosed and tidied up ("this ticket is vague", "fix up ABC-12", "split this, it's too big"). It reads each ticket in full and snapshots it with its last-updated time, says what is wrong with evidence, and proposes a field-by-field before and after change set limited to summary, description, labels, parent, links and new child tickets on a split. It writes nothing until an explicit "apply it" bound to that exact change set, re-checks the ticket for conflicting edits just before writing, never replaces rich descriptions with plain text, never deletes or closes a ticket, and reads each ticket back after applying. Produces the existing and changes blocks of the record.

★ 8