spec
Pass
Audited by Gen Agent Trust Hub on May 10, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the Bash tool to perform Git operations for branch management and version control.
- Evidence: The skill executes
git checkout -b spec/<slug>,git add ralph/specs/<slug>.md, andgit -c commit.gpgsign=false commit -m "spec: <slug>". - Purpose: To create a feature-specific branch and commit the generated markdown specification, facilitating a standard development workflow.
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by ingesting data from the local repository and user input.
- Ingestion points: The skill reads project architecture details from
ralph/AGENTS.md, existing specifications fromralph/specs/, and takes user-provided input viaAskUserQuestion. - Boundary markers: Absent. There are no explicit instructions for the agent to ignore or delimit instructions found within the project's documentation or existing spec files.
- Capability inventory: The skill has the ability to read and write files and execute Bash (Git) commands.
- Sanitization: Absent. The skill does not explicitly sanitize the gathered context before using it to draft the new specification.
Audit Metadata