desk-research-sprint
Pass
Audited by Gen Agent Trust Hub on Oct 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill provides a methodology for the agent to ingest and synthesize information from external sources such as industry reports, forum threads, and documentation. This creates a standard attack surface for indirect prompt injection, though it is inherent to the research task and the skill does not provide tools or scripts for exploitation.\n
- Ingestion points: External websites, reports, and community forums identified in the 'Source Plan' section of SKILL.md.\n
- Boundary markers: None; the capture format does not specify delimiters or warnings for embedded instructions in source material.\n
- Capability inventory: None; the skill is purely instructional and does not contain subprocesses, file-write operations, or network code.\n
- Sanitization: None; the skill relies on the agent's baseline synthesis capabilities.\n- [NO_CODE]: The skill is composed entirely of Markdown instructions and does not ship with any scripts, binaries, or automated tool configurations.\n- [SAFE]: No obfuscation, prompt injection, data exfiltration, or persistence mechanisms were detected. The skill follows best practices for research, including triangulation and confidence labeling.
Audit Metadata