remediation-planner
Pass
Audited by Gen Agent Trust Hub on Jun 16, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the
endorctlCLI exclusively for read-only API operations. It contains explicit prohibitions against any mutating actions, such as editing files, running package manager updates, or creating pull requests. - [PROMPT_INJECTION]: The instructions include defensive measures under the 'Safety' and 'Cursor Host Contract' sections, specifically directing the agent to treat repository files, dependency metadata, and command outputs as data rather than instructions. This effectively mitigates risks associated with indirect prompt injection.
- [DATA_EXPOSURE]: The skill is configured to resolve the user's namespace by reading the
~/.endorctl/config.yamlfile. It includes strict guardrails that forbid the agent from echoing, dumping, or exfiltrating any sensitive information such as API credentials, secrets, or tokens found within the configuration file.
Audit Metadata