Menu
SkillsMP a collecté 17 skills depuis mgrebler/coding-harness. Ouvrez un skill pour examiner sa source et ses détails.
Affichage de 17 skills collectés sur 17.
Validates test files written during the test phase against test-principles.md, spec.md, and constitution.md. Checks task traceability, red-state confirmation, spec coverage, implementation leakage, test isolation, and stack compliance. Returns structured…
Langue du texte source : anglais
Blocking test quality review agent that evaluates test files for assertion quality, test naming discipline, and CI readiness. Produces PASS or FAIL decisions and writes a structured result file for use in autonomous test-phase loops.
Langue du texte source : anglais
Blocking code quality review agent that evaluates implementation quality, maintainability, readability, operational safety, testing discipline, and long-term code health. Produces PASS or FAIL decisions and writes a structured result file for use in…
Langue du texte source : anglais
Validates implemented code against tasks.md, plan.md, spec.md, and constitution.md. Checks TDD order, layer separation, stack constraints, test coverage, contract compliance, and spec adherence. Returns structured pass/fail output. Use after completing a task…
Langue du texte source : anglais
Create or update the feature specification from a GitHub issue's title and body.
Langue du texte source : anglais
Validates plan.md against constitution.md, architecture.md, and spec.md. Use when the user asks to review or critique a plan, check a plan for violations, or run the plan critic.
Langue du texte source : anglais
Validates tasks.md against plan.md, spec.md, and constitution.md. Use when the user asks to review or critique a task list, check tasks.md for violations, or run the tasks critic.
Langue du texte source : anglais
Execute the test phase by writing failing tests for all [TEST] tasks defined in tasks.md. No implementation code is written.
Langue du texte source : anglais
Generate an actionable, dependency-ordered tasks.md for the feature based on available design artifacts.
Langue du texte source : anglais
Runs the automated plan generation and critic loop for the current feature branch by invoking ch_1_plan_auto.py. Handles plan generation, iterative critic review, architecture review, revision, and escalation. Run manually after reviewing the spec.
Langue du texte source : anglais
Runs the automated task generation and critic loop for the current feature branch by invoking ch_2_tasks_auto.py. Handles task generation, iterative critic review, revision, and escalation. Run manually after reviewing the plan.
Langue du texte source : anglais
Runs the automated test phase loop for the current feature branch by invoking ch_3_test_auto.py. Writes failing tests for all [TEST] tasks, runs the iterative test-critic and test-quality-review two-gate loop, applies fixes, and escalates on failure. Run…
Langue du texte source : anglais
Runs the automated implementation and review loop for the current feature branch by invoking ch_4_implement_auto.py. Handles TDD implementation, iterative critic review, code quality review, fixes, and escalation. Run manually after reviewing the test files.
Langue du texte source : anglais
Runs the full plan → tasks → test → implement pipeline for the current feature branch unattended, without stopping for review between stages. Chains ch_1_plan_auto.py, ch_2_tasks_auto.py, ch_3_test_auto.py, and ch_4_implement_auto.py sequentially. Resume-safe…
Langue du texte source : anglais
Blocking architecture review agent that evaluates implementation plans for maintainability, operational safety, scalability, coupling, ownership boundaries, and architectural best practices. Produces PASS or FAIL decisions and writes a structured result file…
Langue du texte source : anglais
Execute the implementation plan by processing and executing all tasks defined in tasks.md
Langue du texte source : anglais
Create or update the feature specification from a natural language feature description.
Langue du texte source : anglais