technical-deslop
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION] (SAFE): The skill uses git diff and git merge-base commands as outlined in references/playbook.md to identify code changes. These are standard, safe operations for its primary purpose.
- [PROMPT_INJECTION] (SAFE): The skill has an indirect prompt injection surface (Category 8) because it processes untrusted data from git diffs. This is inherent to its primary purpose and is mitigated by specific preservation rules.
- Ingestion points: Reads output from git diff commands in references/playbook.md.
- Boundary markers: Absent.
- Capability inventory: Modifies local source files to remove identified slop patterns.
- Sanitization: None, but the playbook and slop-patterns.md provide explicit instructions to preserve security boundaries and avoid unauthorized logic changes.
Audit Metadata