smithery.ai

context-initializer

Automatically invokes init-explorer agent when project context is empty or unknown.

First seen May 4, 2026

Installation

$ npx skills add https://smithery.ai

Similar popular skills

Related neighbors and high-traction skills in the same topics — useful to compare before installing.

Also in this package

Other skills from smithery.ai · top by installs.

npx skills add https://smithery.ai

Browse all from smithery.ai

More details

Agent compatibility

Declared targets from SKILL.md / docs. Unmarked agents are not listed — the skill may still install via the CLI.

Claude Code Declared
Cursor Not declared
Codex Not declared
GitHub Copilot Not declared
Windsurf Not declared
Gemini CLI Not declared
Cline Not declared
OpenCode Not declared

Skill metadata

Parsed from SKILL.md frontmatter.

Declared agents claude-code

Package contents

Files included with this skill beyond the listing page.

  • skill md SKILL.md 2,860 B
  • docs SUMMARY.md 110 B

History

  1. First seen on skills.sh
  2. First recorded snapshot · 1 installs

SKILL.md

Context Initializer

This skill detects when Claude Code lacks project context and automatically invokes the init-explorer agent to gather it.

When to Invoke This Skill

Invoke this skill when ANY of these conditions are true:

  1. No project context: You don't know what this project is about, its tech stack, or its purpose
  2. Missing session history: claude-progress.txt doesn't exist or hasn't been read
  3. Missing architect's digest: architects_digest.md doesn't exist or hasn't been read
  4. User asks context-dependent questions: The user asks about the project but you have no context
  5. Starting a new task: Beginning work on a feature without understanding the codebase

How to Check for Empty Context

Before invoking, verify context is actually missing:

# Check if session files exist
ls -la claude-progress.txt architects_digest.md 2>/dev/null || echo "Context files missing"

Invocation

When context is empty, invoke the init-explorer agent:

Task(subagent_type="init-explorer", prompt="
Gather project context for this codebase.

next_agent: none
task: Explore and document project structure, tech stack, and patterns

After exploration, return a summary of:
1. Project purpose and description
2. Tech stack (languages, frameworks, databases)
3. Key directories and their purpose
4. Testing setup
5. Build/run commands
")

What init-explorer Will Do

The init-explorer agent will:

  1. Orient: Run pwd, ls -la, git log, git status
  2. Read History: Check claude-progress.txt for previous sessions
  3. Read Digest: Check architects_digest.md for task state
  4. Explore: Use Explore subagent for deep codebase analysis
  5. Create Files: Initialize architects_digest.md if missing
  6. Update Progress: Log the exploration session

Expected Output

After init-explorer completes, you will have:

  • Understanding of the project's tech stack and purpose
  • Knowledge of coding patterns and conventions
  • Awareness of test setup and build commands
  • Session logged in claude-progress.txt
  • Task tracking initialized in architects_digest.md

Example Usage

Scenario: User asks "How does authentication work in this project?"

Before: You have no context about the project.

Action: Invoke this skill to run init-explorer.

After: You understand the project uses Flask with JWT authentication, tests are in tests/, and the auth module is in src/auth/.

Do NOT Invoke When

  • You already have project context from earlier in the conversation
  • The user explicitly said not to explore
  • You're in the middle of a task that already has context
  • The init-explorer agent has already run in this session