scheduling-doc-sync
Pass
Audited by Gen Agent Trust Hub on Aug 4, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill implements a robust security model for its GitHub Action workflows, separating AI model execution into read-only, credential-free jobs. All repository mutations are performed by independent, model-free jobs that verify path confinement and file integrity.
- [COMMAND_EXECUTION]: The
scripts/probe-evidence-tool.pyandscripts/bloat-cadence.pyscripts utilizesubprocess.runto execute controlled commands. Execution is limited to a whitelist of tools and restricted to informational flags (--helpor--version), with strict validation of all arguments to prevent command injection. - [EXTERNAL_DOWNLOADS]: The skill clones the vendor's GitHub repository (
github.com/aj604/toolshed.git) to pin the environment to a specific version. This is a standard practice for ensuring workflow stability and security by using verified versions of the plugin tools. - [PROMPT_INJECTION]: The skill processes documentation which represents an indirect prompt injection surface. To mitigate risks,
scripts/render-apply-summary.pyincludes logic to sanitize Unicode control characters and hidden whitespace (e.g., U+2028, U+2029) before they are rendered in pull request summaries or processed by the agent.
Audit Metadata