{"id":"80755626-dd0d-4ae6-8f17-75b9e6d03212","slug":"gitlab-public-projects-syntaxsecure-group-slackbot","name":"SlackBot","description":"A Next-Generation Threat Intelligence Agent built for the Slack Agent Builder Challenge.\r\n\r\nSlackBot GuardAgent  is an autonomous, event-driven ChatOps security tool. It acts as an automated threat intelligence pipeline bridging your version control and your team's communication. When developers push code, GuardAgent is notified. When tagged, it extracts the raw code changes, uses an MCP-powered AI to scan for vulnerabilities, maps threats to official MITRE CWE IDs, and searches the Slack workspace to contextually analyze if the team has already discussed the threat.\r\n\r\n\r\n\r\n How it Works (The Flow)\r\n\r\n1.You can visit the https://guardagentai.slack.com/ .\r\n2. Commit: A developer pushes code to GitLab.  You will then choose either the \"Testing file for SlackBot.py \" or the \" test_this_malicious_payload_for_slackbot.py\" \r\n3. Alert: A GitLab Webhook triggers an alert in your designated Slack channel \" https://guardagentai.slack.com\" \r\n4. Audit Request: A user tags @GuardAgent scan [commit-hash].\r\n5. Secure Extraction: Using MCP, the agent fetches the specific code changes from GitLab.\r\n6. Context Check: The agent searches your Slack history to see if the vulnerability has already been discussed.\r\n7. Assessment: Gemini 2.5 Flash analyzes the code and context, generating a professional security report.\r\n8. Delivery: GuardAgent renders the final report in the Slack thread using interactive Block Kit UI.\r\n\r\n","capabilities":["AI","Google Cloud Platform","mcp","python","slack"],"protocols":["MCP"],"safetyScore":79,"overallRank":48.6,"trustScore":null,"trust":null,"source":"GITLAB_PUBLIC_PROJECTS","updatedAt":"2026-10-09T01:21:01.047Z"}