lovstudio-skill-optimizer
Pass
Audited by Gen Agent Trust Hub on May 7, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands for maintenance workflows, including
git add,git commit,git push, andrsync --delete. These commands are used to synchronize skill files between source and distribution repositories within the vendor's environment. - [PROMPT_INJECTION]: The skill has an indirect prompt injection surface (Category 8) because it reads and processes external data (the content of other skills) to perform optimizations. If a target skill contains malicious instructions, they could influence the agent's behavior during the optimization process.
- Ingestion points: Processes
SKILL.md,README.md, and scripts from the target skill being optimized inStep 3: Apply fixes directly. - Boundary markers: None; the skill does not use delimiters or instructions to ignore embedded prompts in target files.
- Capability inventory: High; the skill has file-write permissions via the Edit tool and command execution capabilities (git, rsync, python3).
- Sanitization: None; content from external skills is used directly to determine and apply fixes.
Audit Metadata