skills/e0ipso/kenkeep/kk-curate/Gen Agent Trust Hub

kk-curate

Pass

Audited by Gen Agent Trust Hub on Jul 29, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill fetches the kenkeep package from the public npm registry (npmjs.com) at runtime using npx kenkeep@latest. This is a well-known service for tool distribution.
  • [REMOTE_CODE_EXECUTION]: Executes a local JavaScript utility, .ai/kenkeep/scripts/kk-detect-root.mjs, to resolve the project's root directory.
  • [COMMAND_EXECUTION]: Invokes several system utilities and tool-specific commands including npx, uuidgen, mkdir, and rm for session management and file cleanup.
  • [PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it ingests session logs from .ai/kenkeep/_sessions/*.md which contain external conversation data.
  • Ingestion points: Reads files matching .ai/kenkeep/_sessions/*.md to extract transcripts.
  • Boundary markers: Absent; there are no specific delimiters or instructions defined to prevent the AI from obeying instructions embedded within the logs.
  • Capability inventory: The skill can perform file writes to the nodes/ directory, execute npx commands, and remove files using rm.
  • Sanitization: Employs npx kenkeep@latest validate to ensure curator outputs conform to the curator-output JSON schema before the persistence phase.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 29, 2026, 03:36 PM
Security Audit — agent-trust-hub — kk-curate