openclaw-executive-assistant-workshop

Pass

Audited by Gen Agent Trust Hub on Jun 14, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions to clone a workshop repository from GitHub (github.com/dandenney/webinars-build-your-own-executive-assistant-with-openclaw.git).
  • [PROMPT_INJECTION]: The skill instructions create a surface for indirect prompt injection by demonstrating how to process untrusted data from external sources, such as files in an 'incoming' folder and exported emails.
  • Ingestion points: Untrusted data enters the agent context via files in code-along/01-data-intake-review/incoming/ and code-along/03-offline-communications-triage/eml/.
  • Boundary markers: The provided prompt templates lack explicit delimiters or instructions to ignore embedded commands.
  • Capability inventory: The workshop patterns involve reading local files and provide bash script examples for automation.
  • Sanitization: No sanitization or validation of the input content is implemented in the prompts.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 14, 2026, 08:47 AM
Security Audit — agent-trust-hub — openclaw-executive-assistant-workshop