synthesis-disclosure-policy
Pass
Audited by Gen Agent Trust Hub on Aug 14, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill defines a set of logical tests and a ledger system to help agents distinguish between public precedents and confidential information. It does not contain executable code, automated shell commands, or network exfiltration logic.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze user-supplied drafts and publications, which represents a potential ingestion surface for indirect prompt injection. 1. Ingestion points: The agent is instructed to process user-provided drafts, edits, and publications in SKILL.md. 2. Boundary markers: Absent; there are no specific instructions to ignore embedded commands within analyzed text. 3. Capability inventory: None; the skill does not request tool access in its frontmatter configuration. 4. Sanitization: Absent; no instructions are provided to filter or escape external content before interpolation. The risk is assessed as safe because the agent lacks the capabilities to perform dangerous actions based on injected instructions.
Audit Metadata