sync-knowledge

Pass

Audited by Gen Agent Trust Hub on Apr 9, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it instructs the agent to ingest data from untrusted sources, such as project documentation (README, docs/) and session logs, and write that data into permanent knowledge base skills.
  • Ingestion points: Project documents, session history, and README files.
  • Boundary markers: Absent; the skill does not specify delimiters to separate untrusted content from the instructions used to update the files.
  • Capability inventory: File system read and write access to the agent's internal configuration directory (~/.claude/skills/).
  • Sanitization: Absent; the skill relies on natural language instructions for the agent to skip secrets, but lacks programmatic sanitization of the input content.
  • [COMMAND_EXECUTION]: The skill performs file system operations (read and write) on the agent's internal configuration directory (~/.claude/skills/). Modifying these files allows for persistent changes to the agent's behavior, which could be exploited to maintain malicious activity across different projects and sessions.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 9, 2026, 05:41 PM
Security Audit — agent-trust-hub — sync-knowledge