cs-doc-sweep
Pass
Audited by Gen Agent Trust Hub on May 20, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes
python .codestable/tools/search-yaml.pyandrg(ripgrep) to catalog and search through project documentation files. These are local operations using vendor-specific tools provided within the project structure.- [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface because it reads and analyzes the content of various project documents to determine if they are obsolete or absorbed by newer designs. Malicious instructions embedded in these documents could potentially influence agent behavior during the analysis phase. - Ingestion points: Reads Markdown and YAML content from
.codestable/features/,.codestable/issues/, and.codestable/refactors/(SKILL.md). - Boundary markers: Absent; the instructions do not specify the use of delimiters or 'ignore' commands when processing document content.
- Capability inventory: The skill has permissions to execute shell commands (
rg), run local Python scripts, and write changes back to the filesystem (updating YAML frontmatter and adding notes). - Sanitization: Absent; content from files is evaluated directly to find evidence for documentation 'sweeps'.
Audit Metadata