copy-editing
Pass
Audited by Gen Agent Trust Hub on Aug 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external text ("existing copy") for editing, which serves as an ingestion point for potentially untrusted data that could contain malicious instructions. The skill provides instructions for PII redaction but lacks explicit boundary markers for input text.\n
- Ingestion points: External copy provided by the user for editing (SKILL.md).\n
- Boundary markers: The instructions do not define specific delimiters to separate untrusted content from the system context.\n
- Capability inventory: The skill utilizes the sequentialthinking and context7 MCP tools and performs terminal execution via the ExecutionProxy Interface (SKILL.md).\n
- Sanitization: The framework includes a security requirement to redact secrets and PII before sharing context with subagents.\n- [COMMAND_EXECUTION]: The skill mandates that terminal actions be executed via the ExecutionProxy Interface (using the
rtkprefix, such asrtk npm test). This establishes a defined pattern for shell command execution within the agent's operational environment.
Audit Metadata