SuperAPI AI Directory
Partner with us中文AI newsFavorites
Skillsspec-test-execute

spec-test-execute

Testing & quality

View repository

Updated 10-16 08:49

About this skill

Spec Test Execution and Reporting (V4: Execute + Report)

This skill produces a "conclusive test report" that answers whether the ACs are met, whether it is deliverable, what the blockers are, and what the next actions are。

Inputs and Gates (must be followed)

REQUIRED SUB-SKILL: currently executing spec-context to obtain context, and echo it back in the conversation FEATURE_DIR=...(allowed (reuse)）。

  • Locate first, then read/write: you must first obtain spec-context to get FEATURE_DIR; if it fails, stop; do not guess paths.
  • Project-level memory is required reading：project/memory/product.md、project/memory/tech.md、project/memory/glossary.md; if it cannot be read, you must write CONTEXT GAP。
  • Input dependencies：verification/test-plan.md and verification/usecase.md must exist;verification/suites.md if it exists, include it in coverage statistics.
  • Defects are not persisted in the Spec Pack: creating verification/bugs/**is prohibited; the report only provides a reference list of "defect ID/link/status/severity".
  • Defect write-back (mandatory): when a defect is created/updated in an external system, it must be written back to this round's report-*.md "Defect List" and linked to the corresponding TC-*。

Output (persisted to disk)

  • {FEATURE_DIR}/verification/report-{date}-{version}.md

Naming rules:

  • date：YYYY-MM-DD
  • version: build/release traceable identifier (version number, build id, git sha); if unknown, use unknown and write in the report CONTEXT GAP: version/build unknown

Mandatory Template (must be followed)

  • report-*.md must be generated according to the template (do not invent structures/field names):
    • assets/report_template.md

Content Requirements (minimum structure)

report-*.md must include at least:

  • Test Summary
    • Conclusion: Pass / Fail / Conditional Pass
    • Version/Build/Environment (if unknown, must write CONTEXT GAP）
  • Coverage Statistics
    • By suite/priority: total, executed, passed, failed, blocked, not executed
  • AC↔TC Coverage Mapping
    • At minimum, it must be able to answer "which ACs are covered by which TCs; which ACs are still gaps (Gap)"
  • Key Failures and Blockers
    • Must be traceable to test case IDs (TC-...)
  • Defect List (references only)
    • Defect system/Issue ID + link + status + severity
    • Must be linked to the corresponding TC-*; if no defect has been filed yet, the blocking reason and next action must be stated (must not be left dangling)
  • Residual Risks and Recommendations
    • Next actions must be executable (rework/retest/canary/monitoring/rollback recommendations)
  • Traceability Links
    • requirements/*、verification/test-plan.md、verification/usecase.md, (optional)verification/suites.md
  • CONTEXT GAP (if any)

DoD Self-Check (V4-DoD)

  • The report has a clear conclusion (not "testing complete")
  • All failures/blockers can be located to TC-* and external defect IDs (if any)
  • Risks and recommendations are executable (next actions are clear)