ghpm-status
Pass
Audited by Gen Agent Trust Hub on Mar 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill is granted
Bash(gh:*)capabilities, allowing it to execute any GitHub CLI command. Although the skill description states it is read-only, the broad permission grant allows for a wide range of interactions with the GitHub API.\n- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it aggregates and processes untrusted data from GitHub project items (such as titles, field values, and custom field options).\n - Ingestion points: Processes data from
.ghpm/config.jsonand cached items from GitHub projects (file:SKILL.md).\n - Boundary markers: There are no explicit instructions to use delimiters or ignore embedded instructions when interpolating project item content into the dashboard output.\n
- Capability inventory: The skill has access to
Bash(gh:*)for GitHub API interaction,Readfor file access, andGrepfor searching (file:SKILL.md).\n - Sanitization: The skill does not define any sanitization or validation logic for the content fetched from the GitHub project before it is displayed or processed.
Audit Metadata