cs-feat-ff
Pass
Audited by Gen Agent Trust Hub on Jun 15, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute a local Python script located at
.codestable/tools/search-yaml.py. This script is used to query project-specific documentation and constraints (learning, decisions, tricks). - [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface because it mandates reading instructions from local project files (e.g.,
.codestable/attention.md) before any other action. - Ingestion points: Files located within the
.codestable/directory. - Boundary markers: None; the agent is instructed to treat the content of these files as authoritative project rules.
- Capability inventory: The skill has the ability to read and write files, execute local Python scripts, and perform Git commits.
- Sanitization: No sanitization or validation of the content of the project files is performed before processing.
- [DATA_EXFILTRATION]: The skill accesses sensitive-appearing directory paths such as
.codestable/and.codestable/architecture/. However, this access is restricted to the local file system for the purpose of maintaining project documentation; no external exfiltration patterns or non-whitelisted domains were identified.
Audit Metadata