ahrefs
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides instructions for a shell command pattern using
curlto interact with the Ahrefs API. This includes a pipeline that utilizesagent-secretsto retrieve a temporary token.- [EXTERNAL_DOWNLOADS]: The skill makes requests toapi.ahrefs.com, which is the official endpoint for the Ahrefs SEO service. These requests are used to fetch keyword research, backlink data, and site metrics.- [DYNAMIC_EXECUTION]: A small Python script is embedded in a shell example to parse JSON data from standard input. This is a local execution used strictly for data processing during the secret retrieval process.- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided inputs such as keywords, domains, and URLs to query the Ahrefs API. This represents an ingestion of untrusted data as part of its core functionality. - Ingestion points: User-supplied strings for keywords and target domains in files like
SKILL.md. - Boundary markers: The skill does not define specific boundary markers or 'ignore' instructions for the interpolated data.
- Capability inventory: Performs network requests via
curland Model Context Protocol (MCP) tool calls. - Sanitization: Relies on the agent's standard tool-calling and parameter handling to format requests correctly.
Audit Metadata