trello-hardened

Pass

Audited by Gen Agent Trust Hub on Apr 21, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes external data from Trello API responses (card names, IDs, descriptions), presenting an indirect prompt injection surface.
  • Ingestion points: Untrusted data enters the agent context via curl command outputs in SKILL.md.
  • Boundary markers: The skill contains a 'Security Guardrails' section with instructions to treat data with caution.
  • Capability inventory: The skill uses curl for network access and jq for data processing, executed via the shell.
  • Sanitization: The instructions explicitly direct the agent to validate IDs and reject shell metacharacters like ;, |, and $().
  • [COMMAND_EXECUTION]: The skill operates by constructing shell commands (curl) with variables. This creates a potential for command injection if input validation is bypassed, a risk the skill author attempts to mitigate through explicit guardrail instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 21, 2026, 06:31 PM
Security Audit — agent-trust-hub — trello-hardened