agent-research-analyst

Pass

Audited by Gen Agent Trust Hub on Jun 12, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection due to its ability to fetch and process untrusted data from the web.
  • Ingestion points: External data is brought into the agent's context through the WebSearch and WebFetch tools.
  • Boundary markers: The skill lacks any instructions or delimiters to isolate or ignore potential malicious prompts embedded in external data.
  • Capability inventory: The agent has access to tools like Read, Write, and Grep for file system interaction.
  • Sanitization: There is no evidence of data sanitization or validation protocols for content retrieved from external sources.
  • [EXTERNAL_DOWNLOADS]: The skill metadata contains an installation command that downloads the skill's source from a well-known public repository (GitHub).
  • [COMMAND_EXECUTION]: The metadata includes shell commands for creating directories and downloading files as part of the setup process.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 12, 2026, 08:18 PM
Security Audit — agent-trust-hub — agent-research-analyst