2718labs/2718lab-devkit/workflow-design
Claude Code and agent skill by 2718labs/2718lab-devkit
- community
Checking that the SKILL.md file is still in the repository.
Reference manual for designing a bounded 2718lab DevKit workflow: scope, dependencies, one-writer boundaries, verification, and completion conditions. Fast Lane is the default policy; this manual does not dispatch sessions or run commands.
Install
npx vskill@latest install 2718labs/2718lab-devkit/workflow-designSKILL.md
Workflow Design Manual
Use Fast Lane as the default planning policy for scoped DevKit work. Record the scope, ambiguity, risk, independent write count, verification cost, blocker, and available capacity before choosing a lane.
For a newly opened project that has no known workspace id or first index snapshot, initialize the local index before deciding that the plugin is degraded: call project_index_register once with the current canonical project root, then call project_index_sync once with the returned opaque workspace id. A successful first snapshot is normal cold start, even when the project already contains source, README, or configuration files. Do not call project_index_status before the first sync and do not report DEGRADED_SKILL_ONLY merely because a new project did not already have an index. Degrade only when the MCP tool is unavailable or register/sync fails; preserve that failure as evidence and do not guess an index identity.
- Keep writers disjoint; prewarm and audit roles are read-only.
- Treat compiler-selected model, effort, worktree, lease, and context as
host-owned facts rather than prose suggestions.
- Refill only after validated terminal evidence; a disconnect or stale context
is not completion.
- Keep temporary files and worktrees inside the approved task root.
This is a planning manual. The bounded compiler and MCP lifecycle contracts are documented in \mcp-tools/devkit_fastlane/\.
Security
- Security scanPassed, 100 of 100Pattern checks on the skill's files.
- Model reviewNot runA model review is scheduled for the next rescan.
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingskills/workflow-design/SKILL.md
Popularity
- Repository stars
- 100
- 7-day trend score
- 0
Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/2718labs/2718lab-devkit/workflow-design)