axiom-apple-docs-research
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill instructions require the agent to execute a shell command (
ls -lt ~/Library/Caches/superpowers/browser/*/session-* | head -5) to identify and navigate to specific session directories within the local file system. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external websites (
developer.apple.com,sosumi.ai) and local browser cache files, creating a vulnerability surface for indirect prompt injection. - Ingestion points: Content is retrieved from external domains via a browser tool and read from markdown files stored in the
~/Library/Caches/superpowers/browser/directory. - Boundary markers: The instructions do not provide specific delimiters or warnings to the agent to treat the captured transcript or documentation content as untrusted data.
- Capability inventory: The agent possesses capabilities for web navigation, shell command execution for file discovery, and local file reading.
- Sanitization: There is no mention of sanitizing, filtering, or validating the external content before it is incorporated into the agent's context.
- [EXTERNAL_DOWNLOADS]: The skill encourages the agent to fetch documentation from
sosumi.ai, a third-party service that serves as a markdown proxy for official documentation, rather than relying exclusively on official sources.
Audit Metadata