lov-agent-instructions
Pass
Audited by Gen Agent Trust Hub on Oct 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local Python scripts (
scripts/profile_store.pyandscripts/validate_skill.py) to manage user profiles and validate skill source integrity. - [INDIRECT_PROMPT_INJECTION]: The skill processes and updates external agent instruction files (e.g.,
AGENTS.md,CLAUDE.md) as its primary workflow, which presents an attack surface for instructions embedded in those files. - Ingestion points: Reads project-specific instruction files and directory structures as specified in the
SKILL.mdworkflow (Step 1). - Boundary markers: The skill lacks explicit boundary markers or instructions to the agent to ignore malicious directives embedded within the analyzed project files.
- Capability inventory: The agent has the ability to write to the file system to update project instructions and manage persistent configuration via
scripts/profile_store.py. - Sanitization: No explicit sanitization or filtering of external instruction content is described before the content is processed or re-written to the file system.
Audit Metadata