write-pr
Pass
Audited by Gen Agent Trust Hub on Jul 10, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the GitHub CLI (
gh) and standard Unix utilities (mktemp,cat,wc) to automate the drafting and editing of Pull Request content. These operations are performed using best practices, such as writing content to temporary files to ensure data integrity. - [SAFE]: The instructions include comprehensive sanitization requirements, directing the agent to strip session-specific metadata, internal abbreviations, and local file paths from the final PR body to prevent information disclosure.
- [SAFE]: Analysis of indirect prompt injection surface (Category 8):
- Ingestion points: The skill processes potentially untrusted external data including git diffs, repository-specific PR templates, and linked Jira or GitHub issue context.
- Boundary markers: The skill does not define specific delimiters for input data but enforces a strict output schema and honorific style.
- Capability inventory: The skill has the capability to modify remote GitHub repositories via
gh pr edit. - Sanitization: It requires the explicit removal of agent-specific branding and conversation-only references from the generated output.
Audit Metadata