Skills for the work you hand off.

Written instructions your bots can load. Skill Guard scans every skill for risky instructions before a bot can use it. That scan doesn’t test the advice itself, so read a skill before you enable it.

Open source. macOS, Windows and Linux.

At a glance

2,243
skills in the library, each scanned
2,203
passed Skill Guard clean. The rest need your OK first
35
subjects, from sales to software to personal finance
  1. You askSay the job in plain words
  2. EmberEmberYour Chief of Staff takes it
  3. A bot with the skillA skillIt opens the skill when the job calls for it
  4. The resultThe finished job

Browse by subject.

Scanned before any bot can use it.

Clean

Ready to use as soon as you add it.

Needs a look

Switches on only after you have read it and said yes.

Dangerous

Blocked.

Skill Guard scans every skill when you import, install, switch on or edit it, and again at startup. Clean skills are ready to use. Anything that needs a look switches on only after you say so, and anything dangerous is blocked. The scan looks for risky instructions, not for good advice: how to run an A/B test, write a cold email or review a contract is still yours to judge. You can also teach a bot a new skill with /learn. (It is on by default; the switch is at the top of Settings → Skills.)

Give your first job to Murage.

Download the app, connect the AI you already pay for, and tell your Chief of Staff what needs doing. Plan on about ten minutes from install to a working team.

No account needed · Runs on the AI plan you already pay for

Search murage.ai

Search the docs, videos, blog, apps, teams, skills and release notes.