slack

Pass

Audited by Gen Agent Trust Hub on Mar 5, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
  • [PROMPT_INJECTION]: The skill exposes a surface for indirect prompt injection by ingesting data from an external source.
  • Ingestion points: The readMessages action in SKILL.md allows the agent to pull untrusted content from Slack channels into its context.
  • Boundary markers: The skill does not define specific delimiters or instructions to treat ingested message content as untrusted data.
  • Capability inventory: The skill includes actions to sendMessage, editMessage, deleteMessage, pinMessage, and react, providing a range of actions an attacker could attempt to trigger.
  • Sanitization: No evidence of content sanitization or validation is present in the tool definition.
  • [NO_CODE]: The skill contains no executable scripts (e.g., Python, JavaScript) or binaries. It consists entirely of a markdown file defining tool parameters and usage examples.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 5, 2026, 02:38 AM