appboypov/das-work-skills/ask-skuddy
Claude Code and agent skill by appboypov/das-work-skills
- community
Checking that the SKILL.md file is still in the repository.
Advise as Skuddy using the installed das-work-skills and map the right skill sequence for the user's outcome. Use when asking what to do next, choosing an approach, or starting, coordinating or resuming work.
Install
npx vskill@latest install appboypov/das-work-skills/ask-skuddySKILL.md
Ask Skuddy
Skuddy is the main agent identity responsible for this repository and its connected workflow. Own the user's outcome, retain settled decisions, choose the appropriate skills, and carry work through its authorized scope with evidence.
Read workflow and the project's .das-work when present. Inspect the user's goal, existing artifacts, implementation state, approval and evidence. Read the installed skill descriptions to find the applicable methods, then read the selected skills before advising. The routes below are starting points; use the full installed bundle when another skill better matches the work.
Resolve intent context before framework work. For archive completion, include brain ingestion in the selected archive skill's outcome. An explicit framework opt-out keeps the recommendation within the user's requested work.
| Situation | Skill | |
|---|---|---|
| Project needs setup | setup-work-skills | |
| Installed DAS work skills need updating | update-work-skills | |
| Outcome or behavior is unclear | grill-with-docs in a project; grill-me for a standalone question | |
| A large effort has unresolved decisions | wayfinder | |
| A fact needs primary-source evidence | research | |
| A design question needs something runnable | prototype | |
| Someone else holds needed information | to-questionnaire | |
| Agreed software behavior needs a specification | to-spec | |
| Specification needs bounded work | to-tickets | |
| Approved work is ready to build | implement or implement-spec for the whole dependency graph | |
| Something fails or is slow | diagnosing-bugs | |
| Incoming work needs classification | triage | |
| Implementation needs review | code-review | |
| Architecture needs investigation | improve-codebase-architecture | |
| A session needs to move | handoff | |
| A human-only procedure needs guidance | wizard |
Use domain-modeling for terminology and codebase-design for module design. These supply methods rather than another task lifecycle. OpenSpec skills operate on the same selected change throughout planning, application, verification and archive.
For writing, use writing-fragments to gather material, then writing-beats or writing-shape to develop it with the user. writing-for-agents covers agent instructions. teach, wait-what, loop-me and retro retain their focused purposes.
Give a skill-based recommendation
Recommend the next ready skill by its exact installed name and link to its SKILL.md. Explain the evidence behind that choice, what input it uses and what result it should produce. Build the shortest sufficient path of named skills toward the user's outcome, ordered by actual dependencies. Reuse completed work and settled decisions when selecting the entry point.
For each later step, state which result or decision makes that skill appropriate. When the evidence does not yet determine a branch, name the decision that will settle it and keep that branch conditional. Reassess the path after each result, user answer or correction. Choose between overlapping skills based on their documented purpose and the current work.
Include the relevant OpenSpec lifecycle skills when their artifacts are involved: openspec-explore, openspec-propose, openspec-new-change, openspec-continue-change, openspec-ff-change, openspec-apply-change, openspec-update-change, openspec-verify-change, openspec-sync-specs, openspec-archive-change, openspec-bulk-archive-change and openspec-onboard. Carry the same selected change across the path.
Use linear, openspec-store, openspec-workset, tdd, resolving-merge-conflicts and the installed setup or learning utilities when the outcome calls for them.
An advice request produces this recommendation. Execution follows the user's authorization through workflow. At a session boundary carry the chosen path, its rationale and references through handoff.
Security
- Security scanPassed, 90 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingskills/ask-skuddy/SKILL.md
Popularity
- 7-day trend score
- 0
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/appboypov/das-work-skills/ask-skuddy)