The Agent Skills Directory

[DATA_EXFILTRATION]: The skill communicates with the official Slack API (api.slack.com) to perform its duties. Sensitive credentials like SLACK_TOKEN and SLACK_WEBHOOK_SECRET are correctly managed through environment variables rather than being hardcoded.\n- [COMMAND_EXECUTION]: Static analysis identified subprocess.run in the test suite (tests/test_slack_cli.py). Review confirms this is used solely to execute the bundled slack-cli script for automated testing purposes and does not pose a security risk.\n- [EXTERNAL_DOWNLOADS]: The slack-cli script is designed to be "stdlib-only," meaning it uses only the Python standard library. It does not download external packages or execute remote scripts at runtime.\n- [INDIRECT_PROMPT_INJECTION]: While the skill processes untrusted Slack messages, it provides explicit instructions to the agent to summarize content and requires human confirmation for any resulting actions, effectively mitigating the risk of instructions embedded in messages being followed by the agent.\n- [SAFE]: The skill implements robust security controls, including human-in-the-loop for write operations and standard HMAC-SHA256 signature verification for webhooks.

slack