knowledge-synthesis
Pass
Audited by Gen Agent Trust Hub on Nov 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes research evidence and existing reports, which are ingestion points for untrusted data.\n
- Ingestion points:
tmp/research/_evidence-{slug}.md,tmp/research/_sources-{slug}.md, and existing reports indocs/knowledge/research/as referenced in the Synthesis and Output sections.\n - Boundary markers: Absent. The instructions do not include specific delimiters or warnings to ignore instructions embedded in the evidence data.\n
- Capability inventory: The skill's capabilities are limited to reading local files (evidence, templates, existing reports) and writing the synthesized report to the
docs/directory. It does not have network access or shell execution capabilities.\n - Sanitization: Absent. There is no requirement to sanitize or validate the content of the evidence files before they are synthesized into the final report.
Audit Metadata