build-graph
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests and parses source files from the repository to construct a knowledge graph.
- Ingestion points: The build_or_update_graph_tool parses files across the repository as described in SKILL.md.
- Boundary markers: No explicit delimiters or instructions to ignore embedded content are defined for the parsing process.
- Capability inventory: The skill utilizes MCP tools to read files and write to a persistent SQLite database located at .code-review-graph/graph.db.
- Sanitization: The documentation does not specify sanitization or filtering of the parsed content.
- Risk: Maliciously crafted comments or code within the repository could be indexed into the graph, potentially influencing future agent reasoning when the graph is queried.
- [SAFE]: The skill relies on pre-defined MCP tools for all operations and manages data locally within the repository structure, which is consistent with its stated purpose of providing code review assistance.
Audit Metadata