slack-standup-digest

Pass

Audited by Gen Agent Trust Hub on Jul 2, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: Potential for indirect prompt injection from untrusted data sources.
  • Ingestion points: The skill reads messages from arbitrary Slack channels (e.g., #engineering, #frontend) using the Slack MCP as described in the Setup Workflow.
  • Boundary markers: The instructions for building the digest prompt in SKILL.md (Step 3) do not specify the use of delimiters or 'ignore' instructions for the ingested Slack message content.
  • Capability inventory: The skill utilizes curl to communicate with the platform's internal automation API (OPENHANDS_HOST) to register tasks.
  • Sanitization: The setup workflow lacks explicit sanitization, validation, or filtering of Slack message content before it is interpolated into the prompt used for digest generation.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 2, 2026, 11:05 AM
Security Audit — agent-trust-hub — slack-standup-digest