self-learning
Pass
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from local project files to build and persist a mental model of the codebase and project conventions.
- Ingestion points: The skill reads various project files including README.md, package.json, directory structures, and representative source code files (Phases 2 and 3).
- Boundary markers: Absent. There are no instructions to the agent to treat file content as data only or to ignore embedded instructions within the analyzed project files.
- Capability inventory: The skill has the capability to execute shell commands for git analysis (Phase 4) and perform file system write operations to the memory/ directory (Phase 5).
- Sanitization: Absent. The skill does not implement validation, filtering, or escaping for content read from external files before it is summarized and stored in the agent's persistent memory.
- [COMMAND_EXECUTION]: The skill utilizes shell commands to extract metadata from the local filesystem and version control system.
- Evidence: Execution of commands such as git log, git shortlog, git branch, ls, and cat (Phases 2 and 4).
Audit Metadata