find-leaks-in-the-wild

Pass

Audited by Gen Agent Trust Hub on Aug 24, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill defines a methodology for the agent to ingest and analyze data from untrusted external sources, such as paste sites, Telegram channels, and leak forums.
  • Ingestion points: External data sources described in SKILL.md and reference/source-catalogue.md (e.g., Telegram public previews, paste aggregators, forum samples).
  • Boundary markers: The instructions do not provide explicit delimiters or instructions for the agent to ignore potentially malicious commands embedded in the processed leak data.
  • Capability inventory: No executable scripts or tool calls are defined within this skill, although it references other skills that may have their own capabilities.
  • Sanitization: The instructions do not specify any sanitization or filtering of the external content before it is processed by the agent.
  • [NO_CODE]: The skill consists entirely of markdown documentation and does not include any executable scripts, configuration files that trigger code execution, or external package dependencies.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 24, 2026, 11:54 AM
Security Audit — agent-trust-hub — find-leaks-in-the-wild