kw-optimize-agents-project
Pass
Audited by Gen Agent Trust Hub on May 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill involves local file system operations, specifically reading, creating, and modifying markdown files within the project directory and the
.planning/subfolder. These actions are aligned with the skill's primary purpose of documentation refactoring. - [PROMPT_INJECTION]: The skill identifies a surface for indirect prompt injection because it ingests and processes content from files such as
AGENTS.mdorCLAUDE.mdwhich could be modified by external actors in a shared repository environment. - Ingestion points: The agent reads existing content from
AGENTS.mdandCLAUDE.mdfor classification and refactoring. - Boundary markers: No explicit boundary markers or instructions to ignore embedded directives within the ingested content are present in the skill instructions.
- Capability inventory: The agent has the ability to read and overwrite local project documentation files.
- Sanitization: There is no evidence of sanitization or filtering of the content being moved from the source files to the new reference documents.
Audit Metadata