iterate-structure-outline
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it ingests and processes untrusted data from multiple external sources. * Ingestion points: The skill reads feedback from conversation history, ticket files, artifact comment blocks, and external files like PRD, TDD, and research documents (SKILL.md Steps 1 & 2, references/feedback-and-comments.md). * Boundary markers: No explicit delimiters or instructions are used to isolate untrusted content from the agent's core instructions. * Capability inventory: The skill can read/write local files, list directories (ls -La), and execute Git commands (SKILL.md Step 6). * Sanitization: There is no evidence of input validation or sanitization for the ingested documentation content.
- [COMMAND_EXECUTION]: The skill executes standard system and version control commands to interact with the local repository. * Evidence: Uses ls -La for directory inventory and multiple Git commands including git rev-parse, git status, and git branch for workspace and branch management (SKILL.md Steps 1 & 6).
Audit Metadata