skills/tbsten/mir/mir-publish-snippet/Gen Agent Trust Hub

mir-publish-snippet

Pass

Audited by Gen Agent Trust Hub on Mar 9, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes multiple shell commands using the npx mir CLI to gather information, publish snippets, and verify installations.
  • [EXTERNAL_DOWNLOADS]: The skill uses npx to execute the mir tool and performs installation tests that fetch snippets from a registry.
  • [PROMPT_INJECTION]: The skill reviews and installs content from potentially untrusted snippets, creating a surface for indirect prompt injection where malicious instructions in the snippet files could influence agent behavior.
  • Ingestion points: Reads snippet.yaml and template files from the target snippet.
  • Boundary markers: No specific delimiters or instructions are used to isolate untrusted instructions within the snippets.
  • Capability inventory: Shell command execution via the npx mir CLI.
  • Sanitization: No explicit sanitization or validation of the snippet content is performed before the agent reviews or installs it.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 9, 2026, 08:25 PM