菜单
Tier-1 orchestrator for running tests and capturing evidence. Executes scenarios, invokes framework runners, collects artifacts, and writes the run + verdict to the ledger. Use when: "run the test", "execute this scenario", "run the suite", "acceptance test this", "capture evidence", "prove it works", "/wicked-testing:execution".
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Evidence-gated acceptance testing with three-agent separation of concerns. Writer designs test plans, Executor collects artifacts, Reviewer evaluates independently. Eliminates false positives from self-grading. Use when: "acceptance test", "verify it works", "did it pass", "run acceptance", "test this scenario", "acceptance criteria", "validate the feature", "/wicked-testing:acceptance"
Record a claim-backing artifact in the vault and attach a deterministic verifier. Use when capturing evidence that "tests pass", "build clean", a commit exists, or a file's contents back a claim — and when replacing stale evidence via supersede. Covers --run vs --artifact, verifier syntax, and contract pinning.
Repo-local structural validator for wicked-testing. Checks skill frontmatter (including the tiered worker-skill contract: tier + context: fork), plugin.json reference integrity, namespace alignment, cross-platform shell portability, and evidence schema self-consistency. Zero LLM cost. Run before publishing or after structural changes. Use when: validating wicked-testing's own skills/plugin.json, pre-publish sanity check, after adding new skills to this repo.
Accessibility specialist — axe-core + pa11y, WCAG 2.1 AA / 2.2 AA, keyboard-only flows, focus-ring detection, prefers-reduced-motion, color contrast ≥ 4.5:1. Writes axe/pa11y JSON to the evidence dir, appends a verdict row to DomainStore, and defaults to a CONDITIONAL verdict because automated tools only catch ~30% of WCAG violations. Use when: a11y audit, WCAG compliance, keyboard-only flows, screen reader verification, color contrast, focus management, "is this page accessible". <example> Context: Reviewer wants a WCAG 2.1 AA check on a new checkout flow. user: "Run an accessibility pass on https://staging.example.com/checkout." <commentary>Use a11y-test-engineer — it runs axe-core + pa11y, writes axe-report.json + pa11y-report.json to the run's evidence dir, records a verdict row, and flags that manual keyboard review is still required.</commentary> </example>
Follows structured wicked-testing test plans step-by-step, collecting evidence artifacts. Executes and captures only — does not judge or grade pass/fail. Writes evidence files to .wicked-testing/evidence/{run-id}/. Use when: acceptance test execution, evidence collection, test plan execution <example> Context: Test plan is ready and needs to be executed step by step. user: "Execute the acceptance test plan for the file upload feature." <commentary>Use acceptance-test-executor for mechanical step execution and evidence capture without judging results.</commentary> </example>
Evaluates evidence artifacts against test plan assertions independently. CRITICAL ISOLATION: Receives ONLY evidence file paths. Never sees execution context. Catches semantic bugs that self-grading misses. Use when: acceptance test review, evidence evaluation, test verdict <example> Context: Executor produced evidence and it needs independent evaluation. user: "Review the evidence from the file upload acceptance tests and render a verdict." <commentary>Use acceptance-test-reviewer for independent, unbiased verdict on test evidence.</commentary> </example>
| name | wicked-testing-execution |
| description | Tier-1 orchestrator for running tests and capturing evidence. Executes scenarios, invokes framework runners, collects artifacts, and writes the run + verdict to the ledger. Use when: "run the test", "execute this scenario", "run the suite", "acceptance test this", "capture evidence", "prove it works", "/wicked-testing:execution". |
| argument-hint | [target] [--project <name>] [--suite] [--json] |
The doer. Takes a scenario or test command, runs it, captures everything,
writes the ledger entry. Evidence lives under
.wicked-testing/evidence/<run-id>/.
/wicked-testing:execution [target] [--project <name>] [--suite] [--json]
target — scenario file path, scenario name, or a test command--project — associate the run with a named project--suite — run the project's full test suite instead of a scenario--json — emit JSON envelope| Input | Dispatch |
|---|---|
| A scenario file | wicked-testing:scenario-executor |
| Scenario + "give me a verdict" (acceptance-grade) | Route to /wicked-testing:acceptance-testing — the 3-role isolated pipeline (writer → executor → reviewer). |
| Scenario + "give me a verdict" (dev-loop, explicit) | wicked-testing:test-designer ONLY if the caller explicitly asks for the fast-path / self-graded loop. |
| Pre-written plan, "just execute" | wicked-testing:acceptance-test-executor |
| "run the suite" (no scenario) | Project's native runner; record result |
| Contract verification | wicked-testing:contract-testing-engineer |
| "wire CI" / "bootstrap CI for tests" | Emit CI trigger (pre-push / GH Actions step); record nothing until a real run executes |
Default posture: verdict requests go to the 3-role pipeline. test-designer
is the dev-loop fast path with known self-grading risk; it is never the default
and never used for audit / CI / crew-phase sign-off evidence. See the warning
in skills/test-designer/SKILL.md.
Every id in the tables above is a forked worker skill (context: fork) —
invoke it with the Skill tool so it runs in an isolated context:
Skill(
skill="wicked-testing:scenario-executor",
args="""Execute the scenario file at the path below and capture evidence.
## Scenario Path
{path to scenarios/<name>.md}
## Evidence Directory
.wicked-testing/evidence/{RUN_ID}/
## Instructions
1. Read the scenario via the Read tool.
2. For each step, run the command via Bash with the scenario's timeout
(enforce via lib/exec-with-timeout.mjs when available — the shell
fallback chain is `timeout || gtimeout || bare` with a warning log).
3. Capture stdout, stderr, exit code, wall-clock duration per step.
4. Write step-N.json + evidence.json + artifact files into EVIDENCE_DIR.
5. Determine per-step outcome: exit 0 = PASS, non-zero = FAIL, CLI missing = SKIPPED.
Do NOT self-grade qualitative outcomes. For acceptance-grade verdicts
route to /wicked-testing:acceptance-testing instead."""
)
Swap the skill id per the table above. For a scenario that also needs
contract verification, dispatch scenario-executor and
contract-testing-engineer in parallel (one Skill(...) call each in the
same turn) and merge results.
For specialized execution paths — chaos experiments, load generators, visual
baselines, etc. — dispatch the specialist. Each writes its own artifacts to
EVIDENCE_DIR and returns an evidence report the skill includes in the run
summary:
| Trigger | Specialist |
|---|---|
| Chaos experiment (Toxiproxy / Chaos Mesh / AWS FIS) | wicked-testing:chaos-test-engineer |
| Load / perf run (k6 / locust / hey) | wicked-testing:load-performance-engineer |
| Visual regression run (Playwright + pixelmatch) | wicked-testing:visual-regression-engineer |
| Full user-journey E2E (multi-context Playwright) | wicked-testing:e2e-orchestrator |
| Component run (RTL + user-event) | wicked-testing:ui-component-test-engineer |
| Integration run (real services via testcontainers) | wicked-testing:integration-test-engineer |
| Fuzz / property run (Hypothesis / fast-check / AFL++) | wicked-testing:fuzz-property-engineer |
| Security run (SAST scan / DAST scan / secrets check) | wicked-testing:security-test-engineer |
| AI-feature test (prompt-injection / eval harness) | wicked-testing:ai-feature-test-engineer |
| IaC validation run (terraform validate / opa / checkov) | wicked-testing:iac-test-engineer |
| Compliance evidence collection (SOC2 / HIPAA controls) | wicked-testing:compliance-test-engineer |
| Selective-execution — "which tests for this diff" | wicked-testing:test-impact-analyzer |
Chaos / load / security-DAST specialists MUST respect the scenario's trust_level frontmatter
field. Production-impacting runs require trust_level: production-authorized
AND a change-ticket: reference; otherwise the specialist refuses and records
SKIP with reason trust-level-insufficient.
run_id (UUID v4 from DomainStore).wicked-testing/evidence/<run-id>/manifest.json is written per docs/EVIDENCE.md (produced by lib/manifest.mjs)wicked.testrun.started,
wicked.test.run.completed, wicked.evidence.captured, and finally
wicked.test.verdict.createdmanifest.json| Old command | Ask execution instead |
|---|---|
run | "run scenario " — execution dispatches scenario-executor and records the run |
ci-bootstrap | "wire CI to run these tests" — execution emits the CI trigger; for the portable evidence gate use wicked-garden's compile |
docs/INTEGRATION.mddocs/EVIDENCE.mdskills/scenario-executor/SKILL.md, skills/test-designer/SKILL.md,
skills/acceptance-test-executor/SKILL.md