CodesearchCAUTION
Multi-repo semantic code search MCP server in Rust — hybrid vector + BM25 retrieval, tree-sitter AST chunking, fully offline. For OpenCode, Claude Code, Cursor, and any MCP client.
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
[](https://opensource.org/licenses/Apache-2.0) [](https://www.rust-lang.org/) [](https://modelcontextprotocol.io/) [](https://github.com/flupkede/codesearch/releases) [](https://github.com/flupkede/codesearch/stargazers)
Multi-repo semantic code search for AI agents — a Rust MCP server with vector + BM25 hybrid retrieval, symbol navigation, and cross-repository orchestration. Fully local, fully offline, no GPU, no Docker.
codesearch gives AI agents (OpenCode, Claude Code, Cursor, and any MCP client) deep codebase understanding through 5 unified MCP tools. Index once, search semantically across multiple repositories simultaneously.
Why codesearch?
- Multi-repo serve mode: Fan-out queries across repository groups with cross-repo RRF ranking
- Hybrid retrieval: Vector embeddings + BM25 full-text search fused with Reciprocal Rank Fusion
- Symbol navigation: Jump to definitions, find usages, trace imports and dependents — in the same tool
- AST-aware chunking: Tree-sitter parsing for 17 languages — chunks align to functions/classes (and Markdown sections), not arbitrary line ranges
- Token-efficient: Returns metadata by default; agents fetch full code only when needed via
get_chunk - Lightweight footprint: Hundreds of MB on disk, runs on CPU only, no runtime model downloads (works behind enterprise proxies)
- Zero config for single repos:
codesearch index && codesearch mcp— done
How does this compare?
The MCP code-search ecosystem grew rapidly in late 2025 / early 2026 and many projects share the same baseline s
9edee2fb3761OBSERVED · 2026-10-07Connect
Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control.
claude mcp add opencode-codesearch-plugin -- npx -y [email protected]
{
"mcpServers": {
"opencode-codesearch-plugin": {
"command": "npx",
"args": [
"-y",
"[email protected]"
]
}
}
}Exposed tools (4)
4 read · 0 write · 0 destructive.
| Tool | Risk | Description |
|---|---|---|
codesearch | read | Semantic code search (codesearch) — query the index and inject the top hits |
codesearch-index | read | Register and index the current directory with codesearch (runs in the background) |
codesearch-status | read | Show codesearch serve health, current project scope and index state |
codesearch_scope | read | Return the codesearch scope for the current directory: project alias, groups, index state, serve health. |
Trust audit
CAUTIONgrade B · trust 83/100 Install with care. The audit found things worth knowing before you trust its output.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | WARN |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | WARN |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- declared (10 observation(s))
- Network
- declared (9 observation(s))
- Shell
- none-observed
- Dependencies
- not all pinned
- Secrets in source
- found
Findings (15)
API_BASE="http://127.0.0.1:${PORT}"if ($env:CODESEARCH_SERVE_PORT) { return "http://127.0.0.1:$($env:CODESEARCH_SERVE_PORT)" }api_key: "supersecret-LEAK-MARKER-do-not-serialize".to_string(),
streamable-http
.codesearchignore
const GREP_GUARD_SH: &str = include_str!("../../integrations/claude-code/hooks/grep-guard.sh");const GREP_GUARD_PS1: &str = include_str!("../../integrations/claude-code/hooks/grep-guard.ps1");const PREAMBLE_SH: &str = include_str!("../../integrations/claude-code/hooks/subagent-preamble.sh");include_str!("../../integrations/claude-code/hooks/subagent-preamble.ps1");const WEB_GUARD_SH: &str = include_str!("../../integrations/claude-code/hooks/web-guard.sh");"url": "http://127.0.0.1:39725/mcp",
`http://127.0.0.1:$CODESEARCH_SERVE_PORT`, else the compiled default
`http://127.0.0.1:39725`). Any HTTP response counts as up and keeps Grep
@types/node, typescript
- **REST `/find-impact` endpoint (HTTP mirror of the `find_impact` MCP tool).** The read-only REST surface (`/search`, `/find`, `/explore`, `/chunk/:id`) now also mirrors `find_impact`: POST a `FindIm
Gates applied: no_behavioural_pass.
9edee2fb3761full audit observations/trust-audit/mcp-server/flupkede__codesearch.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-07 | 9edee2fb3761 | CAUTION | B | 83 | first audit |
Questions
What is the Codesearch MCP server?
Multi-repo semantic code search MCP server in Rust — hybrid vector + BM25 retrieval, tree-sitter AST chunking, fully offline. For OpenCode, Claude Code, Cursor, and any MCP client.
What tools does Codesearch expose?
4 in total: 4 read-only, 0 that write, and 0 that can delete or overwrite. Every one is listed on this page with its risk.
Is Codesearch safe to connect to an agent?
With care. The audit graded it B (83/100) and found 15 things worth knowing before you trust this server, listed below with the exact line each was found on.
What credentials does Codesearch need?
No credential environment variables were found in its source, so it appears to need none.
How does Codesearch run?
It speaks streamable-http, so it runs as a service you connect to over the network. It is published on npm as opencode-codesearch-plugin at 1.0.0.
How current is this page?
The grade is for one exact copy of the source (9edee2fb3761), read on 2026-10-07. The repository is watched and re-audited when it changes.