# Send Gsd Claw to your agent Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually. ## Fast path - Download the package from Yavira. - Extract it into a folder your agent can access. - Paste one of the prompts below and point your agent at the extracted folder. ## Suggested prompts ### New install ```text I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete. ``` ### Upgrade existing ```text I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run. ``` ## Machine-readable fields ```json { "schemaVersion": "1.0", "item": { "slug": "gsd-claw", "name": "Gsd Claw", "source": "tencent", "type": "skill", "category": "效率提升", "sourceUrl": "https://clawhub.ai/RubenAQuispe/gsd-claw", "canonicalUrl": "https://clawhub.ai/RubenAQuispe/gsd-claw", "targetPlatform": "OpenClaw" }, "install": { "downloadUrl": "/downloads/gsd-claw", "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=gsd-claw", "sourcePlatform": "tencent", "targetPlatform": "OpenClaw", "packageFormat": "ZIP package", "primaryDoc": "SKILL.md", "includedAssets": [ "assets/CONTEXT.md", "assets/PLAN.md", "assets/PROJECT.md", "assets/REQUIREMENTS.md", "assets/ROADMAP.md", "assets/STATE.md" ], "downloadMode": "redirect", "sourceHealth": { "source": "tencent", "slug": "gsd-claw", "status": "healthy", "reason": "direct_download_ok", "recommendedAction": "download", "checkedAt": "2026-04-30T03:36:34.282Z", "expiresAt": "2026-05-07T03:36:34.282Z", "httpStatus": 200, "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=gsd-claw", "contentType": "application/zip", "probeMethod": "head", "details": { "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=gsd-claw", "contentDisposition": "attachment; filename=\"gsd-claw-1.0.2.zip\"", "redirectLocation": null, "bodySnippet": null, "slug": "gsd-claw" }, "scope": "item", "summary": "Item download looks usable.", "detail": "Yavira can redirect you to the upstream package for this item.", "primaryActionLabel": "Download for OpenClaw", "primaryActionHref": "/downloads/gsd-claw" }, "validation": { "installChecklist": [ "Use the Yavira download entry.", "Review SKILL.md after the package is downloaded.", "Confirm the extracted package contains the expected setup assets." ], "postInstallChecks": [ "Confirm the extracted package includes the expected docs or setup files.", "Validate the skill or prompts are available in your target agent workspace.", "Capture any manual follow-up steps the agent could not complete." ] } }, "links": { "detailUrl": "https://openagent3.xyz/skills/gsd-claw", "downloadUrl": "https://openagent3.xyz/downloads/gsd-claw", "agentUrl": "https://openagent3.xyz/skills/gsd-claw/agent", "manifestUrl": "https://openagent3.xyz/skills/gsd-claw/agent.json", "briefUrl": "https://openagent3.xyz/skills/gsd-claw/agent.md" } } ``` ## Documentation ### GSD Claw A spec-driven development system adapted for OpenClaw. Transforms vague ideas into verified, working implementations through structured workflows. Based on: glittercowboy/get-shit-done Original Author: Lex Christopherson (@glittercowboy) License: MIT ### When to Use Trigger this skill when user says: "Let's plan [project/feature]" "I want to build [something]" "Scope out [project name]" "GSD mode" or "spec-driven" "Plan and execute [task]" Any substantial project that needs structure NOT for: Quick questions, simple tasks, chat conversations. ### Core Philosophy From the original GSD: "The complexity is in the system, not in your workflow." Principles: Plans ARE prompts — Executable instructions, not documents to interpret Verification built-in — Every task has acceptance criteria Fresh context per task — Subagents prevent context rot Solo developer + AI — No enterprise theater ### Workflow Overview ┌─────────────────────────────────────────────────────────────┐ │ 1. DISCUSS Capture decisions before planning │ │ 2. PLAN Research → Create verified specs │ │ 3. EXECUTE Work through tasks with deviation rules │ │ 4. VERIFY Confirm deliverables actually work │ └─────────────────────────────────────────────────────────────┘ ### Phase 1: DISCUSS Purpose: Capture user's vision BEFORE any planning. Process: Ask about the goal (what are we building?) Identify gray areas based on domain: Visual features → layout, interactions, empty states APIs → response format, error handling Content → structure, tone, depth For each area, ask until user is satisfied Document decisions in {project}/.gsd/CONTEXT.md Output Structure: # Context: [Project/Phase Name] ## Decisions (Locked) - [User decision 1] - [User decision 2] ## Agent Discretion (Freedom Areas) - [Area where agent can choose approach] ## Deferred Ideas (Out of Scope) - [Not doing this now] Rule: Never assume. Always ask. ### Phase 2: PLAN Purpose: Create executable task specs with built-in verification. ### Research (Optional but Recommended) Spawn sub-agent to investigate: Technical approach options Potential pitfalls Best practices for the domain Save to: {project}/.gsd/RESEARCH.md ### Create Plans Each plan = 2-3 tasks maximum (context efficiency). Task XML Structure: Task N: Action-oriented name src/path/file.ts, src/other/file.ts What to do, what to avoid and WHY. Be specific. No guessing. Command or check to prove completion Measurable acceptance criteria Task Types: type="auto" — Agent executes autonomously type="checkpoint:verify" — User must verify type="checkpoint:decision" — User must choose ### Plan Verification Before execution, verify plans against: Tasks are specific and actionable Each task has verify + done criteria Scope matches CONTEXT.md decisions No enterprise bloat Save plans to: {project}/.gsd/plans/{phase}-{N}-PLAN.md ### Phase 3: EXECUTE Purpose: Work through tasks with deviation handling. ### Execution Rules Read plan as literal instructions Execute each task in order Verify done criteria before moving on Apply deviation rules automatically Document everything ### Deviation Rules Apply automatically — track for summary: RuleTriggerActionPermissionBugBroken behavior, errors, security issuesFix → verify → trackAutoMissing CriticalMissing validation, auth, error handlingAdd → verify → trackAutoBlockingPrevents completion (missing deps, wrong types)Fix blocker → trackAutoArchitecturalNew table, schema change, breaking APISTOP → ask userAsk Rule 4 (Architectural) Format: ⚠️ Architectural Decision Needed Current task: [task name] Discovery: [what prompted this] Proposed change: [modification] Why needed: [rationale] Alternatives: [other approaches] Proceed? (yes / different approach / defer) ### Fresh Context Pattern For multi-task execution, spawn sub-agents: Each sub-agent gets fresh 200k context Main context stays lean for user interaction Prevents quality degradation ### Task Completion After each task: Verify done criteria met Commit changes (if applicable) Record in summary Move to next task ### Phase 4: VERIFY Purpose: Confirm deliverables actually work. ### Automated Verification Run verify commands from each task Check files exist Run tests if applicable ### User Acceptance Walk user through testable deliverables: Extract what they should be able to do now Present one at a time Get yes/no or issue description If issues found → create fix plans If all pass: Mark phase complete. If issues: Don't debug manually — create fix plan and re-execute. ### File Structure {project}/ └── .gsd/ ├── PROJECT.md # Vision, always loaded ├── STATE.md # Current position, decisions, blockers ├── REQUIREMENTS.md # Scoped v1/v2 requirements ├── ROADMAP.md # Phases and progress ├── CONTEXT.md # User decisions from discuss phase ├── RESEARCH.md # Domain research (optional) └── plans/ ├── 01-01-PLAN.md # Phase 1, Plan 1 ├── 01-01-SUMMARY.md # Execution results └── ... Templates: Copy from assets/ directory when initializing project files. ### Quick Mode For small tasks that don't need full planning: Trigger: "Quick: [task description]" Process: Skip research and discussion Create single plan with 1-3 tasks Execute with deviation rules Verify and commit When to use: Bug fixes Small features Config changes One-off tasks ### STATE.md Template # Project State ## Current Position Phase: [N] Plan: [N of M] Status: [planning | executing | verifying | blocked] ## Decisions Made - [Decision 1] — [rationale] - [Decision 2] — [rationale] ## Blockers - [Blocker if any] ## Deviations Applied - [Rule N] [Description] — [resolution] Update STATE.md after each significant action. ### Anti-Patterns (Banned) From original GSD: ❌ Enterprise theater Story points, sprint ceremonies Human time estimates (days/weeks) RACI matrices, stakeholder syncs ❌ Vague tasks Add authentication Implement auth ??? ❌ Filler language "Let me...", "I'd be happy to..." "Great question!", "Excellent!" ✅ Direct, imperative, technical ### OpenClaw Adaptations Original GSD → OpenClaw Version GSDOpenClaw/gsd:command slash commandsNatural language triggersClaude Code Task() spawningsessions_spawn sub-agents.claude/ directory{project}/.gsd/ directoryBash scriptsPowerShell/cross-platformgit commit protocolOptional, user preference Preserved Core: XML task structure with verify/done Deviation rules Fresh context via sub-agents State management Plan verification loop ### Example Usage User: "Let's plan the ORO website SEO overhaul" Agent (DISCUSS): Starting GSD workflow for: ORO Website SEO Overhaul Before planning, I need to understand your vision. 1. What's the primary goal? a) Rank higher for specific keywords b) Improve technical SEO scores c) Both d) Something else 2. Any pages that are highest priority? 3. Budget constraints for tools/services? 4. Timeline expectations? After discussion → creates CONTEXT.md Agent (PLAN): ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ GSD ► PLANNING: ORO SEO Overhaul ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ Spawning research sub-agent... [Research complete] Creating plans... - Plan 1: Technical SEO fixes (3 tasks) - Plan 2: Content optimization (2 tasks) - Plan 3: Performance improvements (3 tasks) Verifying plans against requirements... ✓ All checks passed Ready to execute? ### Credits This skill is adapted from Get Shit Done (GSD) by Lex Christopherson (@glittercowboy). Original Repository: https://github.com/glittercowboy/get-shit-done License: MIT Philosophy: Solo developer + AI, no enterprise bloat Thank you Lex for creating an incredibly effective system for building things with AI. ### Version History 1.0.0 — Initial OpenClaw adaptation ## Trust - Source: tencent - Verification: Indexed source record - Publisher: RubenAQuispe - Version: 1.0.2 ## Source health - Status: healthy - Item download looks usable. - Yavira can redirect you to the upstream package for this item. - Health scope: item - Reason: direct_download_ok - Checked at: 2026-04-30T03:36:34.282Z - Expires at: 2026-05-07T03:36:34.282Z - Recommended action: Download for OpenClaw ## Links - [Detail page](https://openagent3.xyz/skills/gsd-claw) - [Send to Agent page](https://openagent3.xyz/skills/gsd-claw/agent) - [JSON manifest](https://openagent3.xyz/skills/gsd-claw/agent.json) - [Markdown brief](https://openagent3.xyz/skills/gsd-claw/agent.md) - [Download page](https://openagent3.xyz/downloads/gsd-claw)