technical-deslop
Pass
Audited by Gen Agent Trust Hub on Oct 3, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes shell commands (
git diff,git merge-base) to identify changes within a repository. These commands are standard for the tool's primary purpose of code maintenance and do not involve unsanitized user input in the command string. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data in the form of git diffs and source code, which could contain adversarial instructions hidden in comments or strings intended to manipulate the agent's cleanup logic.
- Ingestion points: Reads output from
git diff --cachedandgit diff $(git merge-base origin/main HEAD)..HEADas specified inSKILL.mdandreferences/playbook.md. - Boundary markers: The instructions do not define strict delimiters or include explicit warnings to ignore instructions embedded within the content being cleaned.
- Capability inventory: The skill is designed to perform file modifications ("behavior-preserving cleanup") and summarize changes.
- Sanitization: There is no mention of filtering or sanitizing the content of the diffs before processing.
Audit Metadata