codedb-mcp-fast-code-intelligence
Warn
Audited by Gen Agent Trust Hub on Jun 15, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill directs users to clone a repository from an unverified GitHub account (github.com/killop/codedb-mcp.git) to obtain the core software.
- [COMMAND_EXECUTION]: The installation guide instructs the user to compile the Rust project (cargo build --release) and install Node.js dependencies, a process that executes repository-defined build scripts and external packages.
- [REMOTE_CODE_EXECUTION]: The skill sets up an MCP server to execute the resulting local binary, providing a persistent channel for running code from an unverified external source.
- [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface as it processes untrusted codebase data. 1. Ingestion points: Local codebase files indexed by codedb_index and retrieved via codedb_context. 2. Boundary markers: No explicit delimiters or instructions to ignore embedded commands were identified in the tool prompt examples. 3. Capability inventory: The binary performs file system indexing, full-text search, and symbol retrieval, and it is executed persistently as a server. 4. Sanitization: There is no mention of sanitizing or escaping the codebase content before it is passed to the agent for analysis.
Audit Metadata