ai-content-discovery
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill fetches and processes data from external websites (such as robots.txt, llms.txt, and XML sitemaps) using the curl command. This creates a surface where a malicious site could provide data containing prompt injection instructions intended to subvert the agent's logic during verification steps.
- Ingestion points: Output of curl commands in SKILL.md and references/checkpoints.md.
- Boundary markers: None present to distinguish remote content from instructions.
- Capability inventory: The skill utilizes curl, grep, jq, and head for data retrieval and filtering.
- Sanitization: No sanitization or validation of the fetched content is performed before processing.
- [COMMAND_EXECUTION]: The skill relies on shell commands (curl, grep, jq) and a vendor-specific CLI tool (isagentready) for its core verification functionality. These commands are executed to inspect the state of remote web resources.
Audit Metadata