coinagerie-configure

Pass

Audited by Gen Agent Trust Hub on Nov 6, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTIONCOMMAND_EXECUTION
Full Analysis
  • [DYNAMIC_EXECUTION]: The skill instructs the agent to create and modify configuration files (coinagerie.config.ts or variations) that are executed as code by the tool using the jiti library. This allows for arbitrary logic, including the use of Node.js built-in modules like node:fs to read files, to be executed during the tool's initialization phase.
  • [INDIRECT_PROMPT_INJECTION]: The skill describes an attack surface where the tool ingests data from a workspace and performs actions based on it.
  • Ingestion points: Scans files within the working directory based on include and exclude glob patterns (e.g., src/**, scripts/**).
  • Boundary markers: None described in the instructions to delimit or ignore instructions embedded in the scanned files.
  • Capability inventory: The tool can write to the filesystem through the outputs configuration, allowing it to modify files like README.md or generate new JSON glossaries.
  • Sanitization: No sanitization or validation of the scanned file content is mentioned before it is processed into glossary definitions.
  • [COMMAND_EXECUTION]: The instructions reference running the coinagerie CLI with various flags such as --write, --check, and --config to apply configurations and update project documentation.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 6, 2026, 05:15 AM