summarize
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external URLs via WebFetch and local files via Read.
- Ingestion points: The $ARGUMENTS variable takes a URL or file path which is then loaded into the agent's context as described in the "Steps" section of SKILL.md.
- Boundary markers: The instructions lack explicit delimiters or warnings to the agent to ignore instructions embedded within the content being summarized, which increases the risk of the agent obeying instructions found in the source material.
- Capability inventory: The skill has access to Read (file system access), WebFetch (network retrieval), and Write (file creation). An attacker could potentially use indirect injection to trick the agent into reading sensitive files and writing their contents to the workspace summary files.
- Sanitization: There is no evidence of sanitization, filtering, or validation of the fetched content before it is processed by the model.
Audit Metadata