The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a local Node.js script to interact with the Slack API. The command construction in SKILL.md uses environment variables for configuration, which is a standard and safe practice for CLI tools.
[DATA_EXFILTRATION]: While the skill sends data to an external service (Slack), it is restricted to the official https://slack.com/api/chat.postMessage endpoint. The implementation explicitly forbids custom API URLs and third-party token overrides to prevent accidental or malicious data redirection.
[CREDENTIALS_UNSAFE]: The skill handles a SLACK_BOT_TOKEN. Analysis of scripts/slack-post.mjs confirms the token is only read from the process environment and is exclusively used in the Authorization header for the official Slack API. The code contains explicit guardrails to reject tokens passed as command-line arguments, which prevents them from appearing in process logs.
[EXTERNAL_DOWNLOADS]: The script uses native Node.js fetch and does not require or perform any external package installations via npm or other registries.
[PROMPT_INJECTION]: The skill includes safety documentation instructing the agent to only post when explicitly requested and to treat all Slack channels as public/external recipients. It does not contain instructions that override agent safety filters.

slack-post