openspec-codekb-integration

Warn

Audited by Gen Agent Trust Hub on Aug 28, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill requires the installation of an unverified global Node.js package @yun918/codekb (npm install -g @yun918/codekb). This package does not originate from a recognized trusted organization.
  • [INDIRECT_PROMPT_INJECTION]: The skill contains an attack surface where it ingests and processes untrusted data from project documentation.
  • Ingestion points: The tool uses codekb extract to read and distill knowledge from files such as proposal.md, design.md, spec-delta.md, and archive.md (referenced in SKILL.md).
  • Boundary markers: There are no explicit instructions or delimiters defined to prevent the agent from following instructions that might be embedded within these documentation files.
  • Capability inventory: The skill utilizes several tools (codekb_search, codekb_explain, codekb extract) that read from and influence the project's semantic index based on the contents of these files.
  • Sanitization: There is no evidence of sanitization or validation of the natural language content extracted from the specified markdown files before it is used to guide the agent's design decisions.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Aug 28, 2026, 03:54 AM
Security Audit — agent-trust-hub — openspec-codekb-integration