Slk

Read, send, search, and manage Slack messages and DMs via the slk CLI. Use when the user asks to check Slack, read channels or DMs, send Slack messages, search Slack, check unreads, manage drafts, view saved items, or interact with Slack workspace. Also use for heartbeat Slack checks. Triggers on "check slack", "any slack messages", "send on slack", "slack unreads", "search slack", "slack threads", "draft on slack", "read slack dms", "message on slack".

This skill appears to be what it claims: a macOS Slack CLI that auto-extracts your Slack session so it can act as your user. That requires reading sensitive local data (Keychain, Slack Cookies SQLite, LevelDB) and calling local tools (security, sqlite3, openssl, python3, curl). Before installing, consider: 1) Only install on a personal/trusted machine you control — on shared or managed machines this can expose session tokens; 2) Prefer answering Keychain prompts with 'Allow' rather than 'Always Allow' to avoid silent future access; 3) Inspect the npm package (author/repo) and verify its provenance if you require stronger trust; 4) Be aware a token cache (~/.local/slk/token-cache.json) is created — remove it to force re-extraction if needed; 5) If you plan to let an autonomous agent use this skill, understand it can read and send Slack messages as you (highly privileged for personal account actions). If any of the above is unacceptable, do not install or restrict agent permissions.

Type: OpenClaw Skill Name: slk Version: 1.0.0 The skill bundle is classified as suspicious due to its core functionality involving the extraction of sensitive Slack session credentials (xoxc- token and xoxd- cookie) from the macOS Keychain, Slack's SQLite database, and LevelDB files. This is achieved by executing system commands like `security`, `sqlite3`, `openssl`, and `python3` via Node.js `child_process.execSync` and `spawnSync` (primarily in `src/auth.js`). While this capability is high-risk, it is transparently documented in `SKILL.md` and `README.md` as the tool's intended purpose, and there is no evidence of intentional malicious behavior such as unauthorized data exfiltration to external endpoints, persistence mechanisms beyond legitimate caching, or obfuscation. The `SKILL.md` instructions for the AI agent are benign and guide the agent to use the tool for its stated Slack interaction purposes.