Skills · one catalog for Claude Code and Codex
Context Mode keeps your skills on the gateway, so Claude Code and Codex get the same list on any machine. The gateway loads at most one skill a turn, the one that matches your message, and a line in the reply names it.
Paste a GitHub URL or pick a popular repo such as mattpocock/skills. The preview lists each skill with its description and size, and saves nothing.
Add one skill or all of them, and switch any of them off later.
The gateway compares each message with the name and description of every skill you have on. A weak match loads nothing. A match puts the skill's whole body in the prompt, and a line in the reply names it:
⏺ context-mode · skill to-tickets auto-loaded
The skill stays loaded until you say the task is done, another skill loads, or the chat sits idle for 6 hours.
Try it in 5 minutes
You report a real regression in plain words. You never name a skill, yet the right one loads, and when you ask for tickets, a second one usually takes over.
Setup
npx @context-mode/cli # sign in and connect Claude Code, then restart Claude Code
# In the console, open Skills, click the popular repo mattpocock/skills, click Preview import, then Add all.
# Keep clicking "Import the next batch" until all are in. You need diagnosing-bugs and to-tickets on.
git clone https://github.com/sindresorhus/slugify && cd slugify && npm install
npx ava # all green (25 tests): the healthy baseline
python3 -c "p='index.js'; s=open(p).read(); a='|(?:${escapedSeparator})$'; assert a in s; open(p,'w').write(s.replace(a,''))" # play the teammate: a 'tidy up' that drops one branch
git commit -am 'refactor: simplify separator cleanup'
npx ava # now red: 7 tests fail, for example 'Déjà Vu!' gives 'deja-vu-'
claudePrompts
Prompt 1:
Walk me through how slugify turns a title into a slug, step by step.Prompt 2:
Since this morning our blog URLs end with a dash. 'Déjà Vu!' now comes out as 'deja-vu-' and the test suite is red. Nothing in the inputs changed. Figure out what broke and fix it. You can edit the code.Prompt 3:
Good. Now break the follow-up work from this into tickets I can hand to the team.Prompt 4:
Thanks, that's all done.What you will see
# prompt 1: a question about the code, so no skill line
# prompt 2: you never typed a skill name
⏺ context-mode · skill diagnosing-bugs auto-loaded
# prompt 3
⏺ context-mode · skill to-tickets auto-loaded
# prompt 4
⏺ context-mode · skill to-tickets ended, task doneYour team's best habits (debug in order, prove the fix with the tests, split work into clean tickets) apply the moment someone describes the problem in plain words, on every machine, with nothing to install in each repo. What each step does, and how to check it, is in the docs.
Where we stand
Local skills live in folders on each machine or in each repo, so a new laptop starts without your personal skills until you copy them. Context Mode keeps one catalog per account, and Claude Code and Codex both read it through the gateway.
Skills add tokens. Each skill that is on adds one line to a cached list on every request, and a skill that loads adds its whole body. We have not measured whether skills give better answers, so we do not claim they do.
From a public GitHub repo. Paste its URL, preview the skills, and add one or all of them.
No. The gateway compares each message with every skill you have on. Ask for one by name and it loads on that message.
Yes. Each skill that is on adds one line to a cached list, and a skill that loads adds its whole body.
npx @context-mode/cli points Claude Code and Codex at the gateway. Skills are in every plan, and one switch turns them off.