Index Codebases for Agent Search with Code Context Engine
Use Code Context Engine to index a repository and expose token-efficient code search to Claude Code, Codex, Cursor, Gemini CLI, Copilot, OpenCode, and other MCP-capable agents.
npx skills add agentskillexchange/skills --skill index-codebases-for-agent-search-with-code-context-engine
Use Code Context Engine to install the local package, initialize an index for a repository, auto-detect or select the target agent, write the appropriate MCP and instruction config, and route codebase questions through indexed search instead of repeatedly loading large file trees.
What this skill actually does
What the agent/operator does
When to invoke
Invoke this skill when a coding agent is burning context on source discovery, repeatedly rereading files, or needs fast project-aware search across a large repository while preserving local execution and reviewable configuration.
Scope boundary
This is not a generic vector database, MCP server, or token-saving claim. Keep the workflow bounded to repository indexing, agent/editor configuration, and using the generated Code Context Engine search surface for code navigation under a specific project.
Inputs and prerequisites: Python 3.11+, uv or pipx, Code Context Engine CLI, local repository, supported coding agent or MCP-capable editor, optional Ollama/local embedding model.
Setup notes: Run uvx –from “code-context-engine[local]” cce init for one-shot install, index, and agent configuration. For persistent installs, run uv tool install “code-context-engine[local]” , change into the project, and run cce init or cce init –agent codex|claude|all .
Source and verification boundary: use https://elara-labs.github.io/code-context-engine/ as the canonical reference before running the workflow; keep commands, API calls, CLI usage, and generated outputs reviewable against that upstream source.
Framework fit: publish this as a Multi-Framework workflow only when the operator can invoke the documented toolchain directly, rather than treating the upstream project as a generic product listing.