slack-bot-builder
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill describes patterns for processing untrusted user input from Slack messages and slash commands, which creates a potential surface for indirect injection if not handled with care in the final application logic.
- Ingestion points: Event listeners (
@app.message) and command handlers (@app.command) inSKILL.mdingest untrusted content directly from the Slack platform. - Boundary markers: The provided patterns focus on framework usage and do not implement specific boundary delimiters or system instructions to mitigate embedded command execution in processed text.
- Capability inventory: The skill demonstrates the ability to write to channels (
say()) and manipulate UI views (client.views_open()), which are standard capabilities for the described use case. - Sanitization: The patterns utilize standard Slack SDK methods; while they do not include custom sanitization logic, they adhere to the framework's built-in security features such as signature verification.
Audit Metadata