ss-create-pr
Pass
Audited by Gen Agent Trust Hub on Jul 11, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local shell commands including "git" for version control, "gh" for GitHub interactions, and "glab" for GitLab interactions. It also triggers project-specific test runners (e.g., "npm test", "pytest") as part of its pre-flight validation logic.
- [EXTERNAL_DOWNLOADS]: Performs "git fetch" operations to synchronize with remote repositories on GitHub and GitLab. These are standard operations involving well-known technology services.
- [DATA_EXFILTRATION]: Transmits code metadata, commit messages, and branch content to the configured remote repository (GitHub or GitLab) to create pull/merge requests. This behavior is the primary intended function of the skill.
- [PROMPT_INJECTION]: The skill processes commit history and file diffs to generate pull request descriptions. This represents an indirect prompt injection surface where external data influences agent output, although the impact is limited to the content of the created pull request.
- Ingestion points: Commit messages, file diffs, and project templates found in "SKILL.md".
- Boundary markers: None explicitly defined for the generation of PR descriptions.
- Capability inventory: Shell execution for git, gh, and glab commands.
- Sanitization: Relies on standard agent interpolation without specific sanitization for repository content.
Audit Metadata