Skills
skills/sundial-org/awesome-openclaw-skills/jina-reader/Gen Agent Trust Hub

jina-reader

Pass

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: LOWEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION] (LOW): The skill fetches untrusted content from the web, creating a potential surface for indirect prompt injection.
  • Ingestion points: scripts/reader.sh fetches external data via curl from r.jina.ai, s.jina.ai, and g.jina.ai.
  • Boundary markers: Absent; the skill outputs raw content directly to the agent's context without delimiters.
  • Capability inventory: The skill is restricted to curl and jq operations and does not have file-system write or arbitrary code execution capabilities.
  • Sanitization: Absent; the content is passed to the agent without filtering or escaping.
  • [COMMAND_EXECUTION] (LOW): The skill executes curl and jq system binaries to interact with the API. This is expected behavior and safe for the intended use case.
  • [EXTERNAL_DOWNLOADS] (LOW): The skill fetches content from Jina AI endpoints. Although Jina AI is not a predefined trusted source, the risk is categorized as LOW because the fetched content is treated as data for analysis rather than code for execution.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 16, 2026, 07:22 AM