Agent skill

setup-workspace

Configure workspace context and project settings

Stars 163
Forks 31

Install this agent skill to your Project

npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/other/other/setup-workspace-daemn256-workspace-repo-templ

SKILL.md

Setup Workspace

Uses Planner (primary) with Orchestrator support. Guide the creation or update of workspace-specific context. Detects project characteristics and gathers information needed for the agentic kernel to work effectively.

Prerequisites: Access to project root, project files exist (at minimum a README or package manifest)


Modes

Quick Start (Minimum Viable)

For fast setup, only needs:

  1. Project name and purpose
  2. Stack confirmation (auto-detected)
  3. Key documentation pointers

Comprehensive

For full setup, covers all sections in the workspace context schema.


Phase 1: Detection

Scan for Project Markers

  1. Look for README.md
  2. Check for package manifests (package.json, *.csproj, pom.xml)
  3. Look for docs/ or docs/adr/ directories
  4. Scan for configuration files
  5. Report findings

Output

markdown
## Context Anchors

- **Phase:** 1 — Detection

## Detected Markers

| Marker              | Found  | Details         |
| ------------------- | ------ | --------------- |
| README.md           | Yes/No | <path>          |
| Package manifest    | Yes/No | <type and path> |
| docs/ directory     | Yes/No | <structure>     |
| Configuration files | Yes/No | <list>          |

## Next Step

Confirm detection findings are correct.

**Approval Required:** Yes

⛔ CHECKPOINT

STOP. Do not proceed until human explicitly approves:

  • Detection findings are correct
  • No important markers were missed

Phase 2: Stack Inference

Classify Technology

  1. Classify backend technology
  2. Classify frontend technology
  3. Identify documentation structure
  4. Report inference

Stack Categories

Layer Options
Backend .NET / Node / Java / Python / None
Frontend Angular / React / Vue / None
Docs ADR structure / Custom / None

Output

markdown
## Context Anchors

- **Phase:** 2 — Stack Inference

## Inferred Stack

| Layer    | Detection    | Confidence        |
| -------- | ------------ | ----------------- |
| Backend  | <technology> | <High/Medium/Low> |
| Frontend | <technology> | <High/Medium/Low> |
| Docs     | <structure>  | <High/Medium/Low> |

## Next Step

Confirm stack inference is correct.

**Approval Required:** Yes

⛔ CHECKPOINT

STOP. Do not proceed until human explicitly approves:

  • Stack classification is correct
  • Technology choices are accurately detected

Phase 3: Project Identity

Gather Project Information

  1. Ask project name
  2. Ask one-line purpose
  3. Ask about domain terminology
  4. Ask about project-specific conventions

Output

markdown
## Context Anchors

- **Phase:** 3 — Project Identity

## Project Identity

- **Name:** <name>
- **Purpose:** <one-line purpose>
- **Domain terms:** <key terms>
- **Conventions:** <project-specific conventions>

## Next Step

Continue to preview.

**Approval Required:** No

Phase 4: Preview and Confirm

Generate and Present

  1. Generate workspace context from gathered information
  2. Present preview of the full context file
  3. Apply after approval

Workspace Context Schema

The workspace context file contains:

  • Project identity (name, purpose)
  • Domain terminology
  • Architecture (pattern, backend, frontend)
  • Convention amendments (project-specific overrides)
  • Key documentation pointers

Output

markdown
## Context Anchors

- **Phase:** 4 — Preview and Confirm

## Generated Workspace Context

<full preview of docs/workspace/context.md content>

## Next Step

Awaiting approval to write workspace context file.

**Approval Required:** Yes

⛔ CHECKPOINT

STOP. Do not write the workspace context file until human explicitly approves:

  • Content is accurate
  • No sections are missing
  • Terminology is correct

Error Handling

Error Recovery
Empty project (no markers) Ask user to describe project manually
Conflicting markers Present all detected, ask for clarification
Existing context file Offer to update vs replace

Expand your agent's capabilities with these related and highly-rated skills.

Didn't find tool you were looking for?

Be as detailed as possible for better results