cs-explore
Pass
Audited by Gen Agent Trust Hub on Jun 24, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses a local Python utility located at
.codestable/tools/search-yaml.pyto filter metadata and check for duplicate exploration documents. This involves executing shell commands with arguments derived from the skill's state. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection. It is designed to ingest untrusted data from the repository's source code using discovery tools like glob, grep, and read operations.
- Ingestion points: Reads arbitrary files from the local repository during the Phase 2 exploration stage.
- Boundary markers: The instructions do not define delimiters or specific warnings to ignore instructions embedded within the codebase being analyzed.
- Capability inventory: The agent can write files to the
.codestable/compound/directory and execute the localsearch-yaml.pyscript. - Sanitization: There is no evidence of sanitization or filtering of the content retrieved from the codebase before it is used to formulate conclusions or document summaries.
Audit Metadata