results-report
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external files, which introduces a surface for indirect prompt injection if the source data is maliciously crafted.
- Ingestion points: The skill reads multiple local artifacts including
analysis-report.md,stats-appendix.md,figure-catalog.md, and various reference files. - Boundary markers: The instructions include strict constraints called "Hard gates" that prevent execution if specific files are missing. It also explicitly requires preserving original claim wording from source artifacts to prevent over-interpretation.
- Capability inventory: The skill has the ability to write or update markdown files in several directories (
Results/Reports/,Results/,Daily/) and update the.claude/project-memory/file. - Sanitization: No specific sanitization or filtering of the input text is mentioned; the skill relies on instructional constraints to guide the interpretation of findings.
Audit Metadata