environmental-analysis
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill researches environmental data by fetching content from external websites via the WebFetch tool. This creates a surface for indirect prompt injection where malicious instructions embedded in third-party websites could attempt to influence the agent's behavior.
- Ingestion points: External content is retrieved through WebFetch during the research workflow described in SKILL.md.
- Boundary markers: The skill includes instructions to treat source content as evidence rather than authority to modify other records or broaden the task, but it lacks physical delimiters or markers for the processed external data.
- Capability inventory: The agent is granted access to the Bash, Write, Edit, and Read tools, as well as process.run operations defined in the host contract.
- Sanitization: There is no explicit logic to sanitize or escape fetched content before it is interpolated into reports or potential tool command contexts.
- [COMMAND_EXECUTION]: The skill's configuration in SKILL.md and host-contract.json explicitly allows the use of the Bash tool and process.run operations. While the intended use is for research and file management, this provides a powerful execution environment that could be misused if the agent is compromised by malicious input.
Audit Metadata