docs-check-contradictions

Pass

Audited by Gen Agent Trust Hub on Aug 12, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it ingests and analyzes documentation content from external files and web pages.
  • Ingestion points: The skill reads local markdown files provided in arguments and fetches remote documentation content using the WebFetch tool or MCP server tools (SKILL.md).
  • Boundary markers: None identified. There are no instructions or delimiters used to prevent the agent from potentially executing instructions hidden within the ingested documentation.
  • Capability inventory: The skill uses the Read, Grep, Glob, and WebFetch tools, providing it with file system and network access.
  • Sanitization: None identified. The skill extracts factual assertions directly from the text without sanitizing the inputs.
  • [EXTERNAL_DOWNLOADS]: The skill fetches documentation from https://www.elastic.co/docs/. This is a trusted vendor resource for the author 'elastic' and the activity is legitimate for the tool's intended purpose.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 12, 2026, 07:02 AM
Security Audit — agent-trust-hub — docs-check-contradictions