skills.rest
JPFrancoia
JPFrancoiaCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

codecompanion-memory

Retrieve past CodeCompanion conversations from a local ChromaDB via semantic search.

Updated Dec 15, 2025
Search Tags:#context retrieval#semantic search#memory#conversation history#ChromaDB#vectorcode
One-click install
npx skills add https://github.com/JPFrancoia/skills --skill codecompanion-memory
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: codecompanion-memory
Source: https://github.com/JPFrancoia/skills/tree/main/codecompanion-memory
Command: npx skills add https://github.com/JPFrancoia/skills --skill codecompanion-memory

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires vectorcode.

What problem does it solve?

This Skill provides a fast, semantic way to locate relevant past CodeCompanion conversations stored in a local vector database. It helps you recall decisions, patterns, and solutions without manual search.

Core Features & Use Cases

  • Semantic search across conversations: Find relevant past discussions based on meaning, not just keywords.
  • Context retrieval for debugging and architecture: Revisit previous designs to inform current work.
  • Portable history access: Works with local history stored in ChromaDB for Claude Code users.

Quick Start

Run a sample search with the provided script, for example: ~/.claude/skills/codecompanion-memory/query.sh --query "authentication issue" --count 3

Frequently Asked Questions about codecompanion-memory

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I search past CodeCompanion conversations by meaning rather than keywords?▼

Semantic search across conversations retrieves relevant past discussions based on meaning using embeddings stored in a local ChromaDB vector database. Run the query script with a natural language search term to find related solutions, decisions, or code samples without exact keyword matching.

Can I retrieve conversation history for debugging without manual searching?▼

Yes. The Skill queries a local vector database to retrieve relevant past CodeCompanion conversations instantly. Provide a --query parameter describing your problem, and it returns matching historical context to inform current debugging or architecture work.

What do I need to set up before using semantic search on my conversation history?▼

You need VectorCode as a dependency and a local ChromaDB instance storing your CodeCompanion conversation history. The Skill provides a Bash CLI wrapper that queries this local vector database using semantic embeddings.

How do I control the number of past conversations returned in a search?▼

Use the --count parameter when running the query script to specify how many relevant conversations to retrieve. Additional options include --verbose for detailed output and --project to scope searches to a specific project.

Can I access my CodeCompanion history portably across different environments?▼

The Skill works with conversation history stored locally in ChromaDB, making it portable across Claude Code environments where that history exists. The vector database remains on your machine for fast semantic retrieval.

What's the difference between keyword search and semantic search for finding relevant past code discussions?▼

Keyword search matches exact terms, while semantic search finds conceptually related conversations even with different wording. This Skill uses semantic embeddings to surface solutions addressing the meaning of your query, not just matching phrases.

Related Skills

Explore More Data & Analytics Skills→
jonnabiojonnabio
6

chroma

Index and retrieve documents using semantic embeddings and metadata.

Community
Intermediate
xurenluxurenlu
1

remembering-conversations

Search and synthesize past conversations for relevant context and solutions.

Community
Intermediate
adeboldadebold

AgentDB Vector Search

Perform vector-based semantic search against an AgentDB index.

Community
Advanced
BEKO2210BEKO2210
15

search

Search knowledge bases semantically with vector embeddings and keyword fallback.

Community
Advanced