How do I install the Gsd Validation skill?

Run npx skillmds@latest add digi4care/gsd-validation 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 Gsd Validation 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 Gsd Validation?

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 Gsd Validation free to use?

Yes. Installing skills from SkillMD is free. This skill is licensed under MIT.

Who published Gsd Validation?

digi4care (@digi4care) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Gsd Validation by digi4care — Agent Skill

Name: Gsd Validation
Author: digi4care

File contents

gsd-validation

Automated post-phase validation workflow for GSD phases.

Purpose

Run a deterministic verification pipeline after implementation:

verify work
analyze flow gaps
apply HITL decision gate when gaps exist
execute runtime test stack (servers, browser, logs)
emit final report artifacts

Triggers

Use this skill when the request includes:

validating a finished phase end-to-end
combining /gsd-auto-validate and /gsd-analyze-flow
automated browser + server log validation
safe teardown requirements in shared/swarm environments

Workflow

Parse phase id and execution mode (quick|full)
Run /gsd-auto-validate <phase>
Run /gsd-analyze-flow <phase>
If gaps exist, enforce HITL gate using references/gate-policy.md
If approved for tests, execute runtime tools in this order:
- gsd-run-create
- gsd-server-start
- gsd-server-wait-ready
- gsd-playwright-execute
- gsd-logs-collect
- gsd-server-stop-owned
- gsd-run-finalize
Render final report from references/report-template.md

Runtime Tool Contract Expectations

gsd-run-create: creates run workspace and returns runId
gsd-server-start: starts tracked process and records ownership metadata
gsd-server-wait-ready: verifies readiness with probe evidence
gsd-playwright-execute: runs browser tests and stores artifacts
gsd-logs-collect: produces structured issues with severity
gsd-server-stop-owned: terminates only validated run-owned processes
gsd-run-finalize: emits FINAL-VALIDATION.md and FINAL-VALIDATION.json

Failure Handling

On verify or flow-analysis failure: finalize run as failed
On gap gate stop or fix-plan path: finalize run as blocked
On runtime tool failure: continue teardown, then finalize run as failed
On teardown safety skip: include warnings in final report, never force kill unknown processes

Modes

Mode	Behavior
quick	smoke-level test depth, minimal artifacts
full	complete flow checks, visual evidence, extended logs

Safety Rules

never kill processes not proven as run-owned
never bypass gap gate without explicit user decision
always preserve run artifacts even on partial failure

References

references/gate-policy.md
references/report-template.md

---name: gsd-validationdescription: Automated post-phase validation workflow for GSD phases with deterministic gates and HITL controls.license: MIT---# gsd-validationAutomated post-phase validation workflow for GSD phases.## PurposeRun a deterministic verification pipeline after implementation:1. verify work2. analyze flow gaps3. apply HITL decision gate when gaps exist4. execute runtime test stack (servers, browser, logs)5. emit final report artifacts## TriggersUse this skill when the request includes:- validating a finished phase end-to-end- combining `/gsd-auto-validate` and `/gsd-analyze-flow`- automated browser + server log validation- safe teardown requirements in shared/swarm environments## Workflow1. Parse phase id and execution mode (`quick|full`)2. Run `/gsd-auto-validate <phase>`3. Run `/gsd-analyze-flow <phase>`4. If gaps exist, enforce HITL gate using `references/gate-policy.md`5. If approved for tests, execute runtime tools in this order: - `gsd-run-create` - `gsd-server-start` - `gsd-server-wait-ready` - `gsd-playwright-execute` - `gsd-logs-collect` - `gsd-server-stop-owned` - `gsd-run-finalize`6. Render final report from `references/report-template.md`## Runtime Tool Contract Expectations- `gsd-run-create`: creates run workspace and returns `runId`- `gsd-server-start`: starts tracked process and records ownership metadata- `gsd-server-wait-ready`: verifies readiness with probe evidence- `gsd-playwright-execute`: runs browser tests and stores artifacts- `gsd-logs-collect`: produces structured issues with severity- `gsd-server-stop-owned`: terminates only validated run-owned processes- `gsd-run-finalize`: emits `FINAL-VALIDATION.md` and `FINAL-VALIDATION.json`## Failure Handling1. On verify or flow-analysis failure: finalize run as `failed`2. On gap gate stop or fix-plan path: finalize run as `blocked`3. On runtime tool failure: continue teardown, then finalize run as `failed`4. On teardown safety skip: include warnings in final report, never force kill unknown processes## Modes| Mode | Behavior || ----- | ---------------------------------------------------- || quick | smoke-level test depth, minimal artifacts || full | complete flow checks, visual evidence, extended logs |## Safety Rules- never kill processes not proven as run-owned- never bypass gap gate without explicit user decision- always preserve run artifacts even on partial failure## References- `references/gate-policy.md`- `references/report-template.md`

Frequently asked questions

Gsd Validation