kata-technical-writing
Pass
Audited by Gen Agent Trust Hub on Nov 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests 'source-material' and 'readme-brief' from external sources to drive documentation generation and verification tasks. This creates a surface where instructions embedded within these source materials could attempt to influence the agent's behavior or output. Evidence:
- Ingestion points: 'SKILL.md' consumes 'source-material' and 'technical-writing-request'.
- Boundary markers: The instructions do not explicitly direct the agent to ignore or delimit instructions found within the ingested source material.
- Capability inventory: The skill permits local file edits and the execution of verification commands (e.g., 'mix docs' or link checkers).
- Sanitization: No specific sanitization or escaping mechanisms for the ingested text are mentioned.
- [COMMAND_EXECUTION]: The skill authorizes the agent to execute local commands for verifying the accuracy of documentation, such as testing code examples, checking links, and running documentation build tools. While the skill advises using the 'narrowest available check' and discourages installing new tools, the specific commands to be run are influenced by the content of the documentation being processed.
Audit Metadata