results-section-writer
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user data, creating a surface for indirect prompt injection attacks where malicious instructions could be embedded in research summaries. 1. Ingestion points: As detailed in SKILL.md, the agent accepts Results outlines, figure inventory, analysis summaries, and partial manuscript drafts. 2. Boundary markers: The skill relies on natural language instructions in 'references/results-boundary-rules.md' and 'Hard Rules' to maintain evidence boundaries, but it lacks specific technical delimiters or structural markers to distinguish user data from system instructions. 3. Capability inventory: The skill is a text-generation tool (Mode A) and does not have access to tools for network communication, shell access, or file system modifications. 4. Sanitization: No explicit sanitization or filtering of external input is implemented within the provided reference modules.
- [NO_CODE]: The skill is entirely instruction-based and does not contain, generate, or execute any programming code or scripts.
- [SAFE]: The external repository link (github.com/aipoch) points to the author's own official source and is classified as a legitimate vendor resource without security risk.
Audit Metadata