slack-standup-digest

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). [SKILL.md, Step 1/3/4: the required workflow uses the Slack MCP to “search for recent messages in one of the source channels” and then constructs a digest prompt that will include those Slack messages, so outsider-authored Slack free text from user-selected channels is ingested at runtime.]

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.80). The skill issues a runtime curl POST to "${OPENHANDS_HOST}/api/automation/v1/preset/prompt" to register a prompt preset (and thereby trigger remote automation execution), so it depends on an external runtime service that will store and run prompts.