knowledge-ops

Pass

Audited by Gen Agent Trust Hub on Sep 19, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is primarily designed to ingest and structure data from external, untrusted intelligence sources (OSINT, social intelligence, and deep research).
  • Ingestion points: Intelligence gathered from external tools and research skills is written to markdown files in the memory/ directory and subsequently processed by internal Bash and Python aggregation scripts.
  • Boundary markers: The skill uses standard YAML frontmatter to separate metadata from content, but it does not implement explicit boundary markers or instructions to the agent to ignore or isolate potentially malicious commands embedded in the processed intelligence.
  • Capability inventory: The skill utilizes Bash for command execution and script generation, Write and Edit for knowledge base maintenance, and WebFetch for external API communication.
  • Sanitization: The provided utility scripts for entity extraction and reporting rely on regular expressions and basic string manipulation. They do not feature robust sanitization to prevent the agent from misinterpreting embedded instructions as its own commands.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 19, 2026, 04:36 PM