How do I install the Quality Gates skill?

Run npx skillmds@latest add microsoft/quality-gates in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Quality Gates safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Quality Gates?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Quality Gates free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Quality Gates?

Microsoft (@microsoft) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Quality Gates by microsoft — Agent Skill

Name: Quality Gates
Author: Microsoft

Quality Gates

Runs quality gate validation at each workflow stage. Supports 4 gate types: spec-quality, spec-to-plan, plan-to-tasks, completeness. Produces gate pass/fail reports with actionable feedback. Triggers: "run quality gate", "validate spec quality", "spec quality", "plan coverage", "check plan coverage", "validate plan to tasks", "check task traceability", "verify task completeness", "run completeness review", "review implementation", "check plan traceability", "check coverage", "completeness quality gate", "final sign-off", "feature parity sign-off", "completeness gate", "feature parity", "sign-off report", "verify all requirements". Requires `type` parameter to select the gate. NOT for: code review, implementation (use implementing-code), spec writing.

Microsoft 87372fc 8 files · 30.2 KB Updated Sep 17, 2026 2.7k repo stars

File contents

User Input

You MUST consider the user input before proceeding (if not empty).

Output

Reports and checkpoints are written under your task's Artifact path: (from task metadata).

checklists/requirements.md           ← spec-quality gate
checkpoints/spec-to-plan.yaml        ← spec-to-plan gate
traceability-matrix.yaml             ← completeness gate
migration-summary.md                 ← completeness gate

Input

type = spec-quality | spec-to-plan | plan-to-tasks | completeness

Workflow

Step 1: Load Context

Load from cwd:

constitution.md — non-negotiable; violations are always CRITICAL
spec.md, plan.md, tasks.md, checkpoints/ (as relevant for the gate type)

Step 2: Gate Routing

Load the checklist reference for the selected gate:

type	Reference	Key Output
`spec-quality`	`references/gate-spec-quality.md`	`checklists/requirements.md`
`spec-to-plan`	`references/gate-spec-to-plan.md`	`checkpoints/spec-to-plan.yaml`
`plan-to-tasks`	`references/gate-plan-to-tasks.md`	`checkpoints/plan-to-tasks.yaml`
`completeness`	`references/gate-completeness.md`	traceability-matrix.yaml + migration-summary.md

Step 3: Execute Gate

Follow the process defined in the referenced gate file.

Step 4: Verdict

PASS: all checklist items pass → ✓ [gate] PASSED.
FAIL: any CRITICAL → ✗ [gate] FAILED. N critical issues. + remediation steps

Rules

READ-ONLY: never modify source code; only write checklists, checkpoints, reports.
Max 50 findings per gate.
Deterministic: rerunning without changes = same results.
Never hallucinate: if artifact is missing, report it.

Severity

Level	Meaning
CRITICAL	Checkpoint failed, broken traceability, constitution violation, requirement not implemented, testing strategy deviation without documented blocker
HIGH	Partial requirement, implementation drift, untested acceptance criteria
MEDIUM	Incomplete non-functional coverage, terminology drift
LOW	Style/naming, documentation gaps

Resources

references/gate-spec-quality.md — spec quality checklist and process
references/gate-spec-to-plan.md — spec-to-plan checklist and process
references/gate-plan-to-tasks.md — plan-to-tasks checklist and process
references/gate-completeness.md — completeness checklist and process (includes plan coverage)
templates/report-template.md — migration summary structure

---name: quality-gatesdescription: Runs quality gate validation at each workflow stage. Supports 4 gate types: spec-quality, spec-to-plan, plan-to-tasks, completeness. Produces gate pass/fail reports with actionable feedback. Triggers: "run quality gate", "validate spec quality", "spec quality", "plan coverage", "check plan coverage", "validate plan to tasks", "check task traceability", "verify task completeness", "run completeness review", "review implementation", "check plan traceability", "check coverage", "completeness quality gate", "final sign-off", "feature parity sign-off", "completeness gate", "feature parity", "sign-off report", "verify all requirements". Requires `type` parameter to select the gate. NOT for: code review, implementation (use implementing-code), spec writing.---## User InputYou **MUST** consider the user input before proceeding (if not empty).## OutputReports and checkpoints are written under your task's `Artifact path:` (from task metadata).```checklists/requirements.md ← spec-quality gatecheckpoints/spec-to-plan.yaml ← spec-to-plan gatetraceability-matrix.yaml ← completeness gatemigration-summary.md ← completeness gate```## Input`type` = `spec-quality` | `spec-to-plan` | `plan-to-tasks` | `completeness`## Workflow### Step 1: Load ContextLoad from cwd:- constitution.md — **non-negotiable**; violations are always CRITICAL- spec.md, plan.md, tasks.md, checkpoints/ (as relevant for the gate type)### Step 2: Gate RoutingLoad the checklist reference for the selected gate:| type | Reference | Key Output ||------|-----------|------------|| `spec-quality` | `references/gate-spec-quality.md` | `checklists/requirements.md` || `spec-to-plan` | `references/gate-spec-to-plan.md` | `checkpoints/spec-to-plan.yaml` || `plan-to-tasks` | `references/gate-plan-to-tasks.md` | `checkpoints/plan-to-tasks.yaml` || `completeness` | `references/gate-completeness.md` | traceability-matrix.yaml + migration-summary.md |### Step 3: Execute GateFollow the process defined in the referenced gate file.### Step 4: Verdict- **PASS**: all checklist items pass → `✓ [gate] PASSED.`- **FAIL**: any CRITICAL → `✗ [gate] FAILED. N critical issues.` + remediation steps## Rules- **READ-ONLY**: never modify source code; only write checklists, checkpoints, reports.- **Max 50 findings** per gate.- **Deterministic**: rerunning without changes = same results.- **Never hallucinate**: if artifact is missing, report it.## Severity| Level | Meaning ||-------|---------|| CRITICAL | Checkpoint failed, broken traceability, constitution violation, requirement not implemented, **testing strategy deviation without documented blocker** || HIGH | Partial requirement, implementation drift, untested acceptance criteria || MEDIUM | Incomplete non-functional coverage, terminology drift || LOW | Style/naming, documentation gaps |## Resources- `references/gate-spec-quality.md` — spec quality checklist and process- `references/gate-spec-to-plan.md` — spec-to-plan checklist and process- `references/gate-plan-to-tasks.md` — plan-to-tasks checklist and process- `references/gate-completeness.md` — completeness checklist and process (includes plan coverage)- `templates/report-template.md` — migration summary structure

Frequently asked questions