thinking-redteam
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's core function involves processing and analyzing untrusted content (ideas, plans, arguments) which can be provided directly or fetched via WebFetch. This creates a surface for indirect prompt injection if the analyzed content contains hidden instructions.
- Ingestion points: The skill instructs the agent to analyze specific arguments or plans provided in the input or retrieved from the web.
- Boundary markers: There are no instructions to use delimiters or to treat the analyzed content as data only, lacking protection against embedded commands.
- Capability inventory: The agent has access to file system tools (Read, Grep, Glob) and network tools (WebSearch, WebFetch), which could be misused if the agent obeys instructions found within the analyzed text.
- Sanitization: The skill does not provide methods for escaping or validating the content before it is processed by the agent.
Audit Metadata