octocode-documentation-writer

Pass

Audited by Gen Agent Trust Hub on Mar 11, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill architecture is susceptible to indirect prompt injection as it processes untrusted source code from external repositories. \n
  • Ingestion points: File contents and directory structures are read across multiple phases, including Discovery (Phase 1), Research (Phase 3), Writing (Phase 5), and QA (Phase 6). \n
  • Boundary markers: The skill does not implement explicit delimiters or instructions for sub-agents to ignore potentially malicious content within the analyzed code. \n
  • Capability inventory: The pipeline has file read/write access and the ability to spawn parallel sub-agents with broad tool access. \n
  • Sanitization: No explicit sanitization or instruction-filtering is applied to the repository data before it is processed by the AI agents. \n- [EXTERNAL_DOWNLOADS]: The skill requires the installation of octocode-cli and octocode-mcp via npx, which are vendor-owned resources (bgauryy) necessary for the skill's functionality. \n- [COMMAND_EXECUTION]: The orchestration logic executes various local discovery and search operations using specialized MCP tools to understand the project structure and logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 11, 2026, 10:17 PM