connect
Skill27 starsChanged 3 months agoMITView on GitHub
Connect this coding agent installation to the user's Altitude account via the device pairing flow.
- Installs packages
64 other projects use a version of this.
Adopted by
Most recently taken up by
- thensls/nsls-builder-toolkittheir own version, first seen Nov 8, 2026
- langchain-ai/open-swetheir own version, first seen Nov 8, 2026
- BrightGold70/skillstheir own version, first seen Nov 8, 2026
- flung-official/flung-claude-plugintheir own version, first seen Nov 8, 2026
- matthewwong525/WongStacktheir own version, first seen Nov 8, 2026
- CraftOS-dev/Agent-Apptheir own version, first seen Nov 8, 2026
Used with
Projects with it also use connect-apps, skill-share and 10 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
ArogyaReddySkillconnect-apps
Connect Claude to external apps like Gmail, Slack, GitHub. Use this skill when the user wants to send emails, create issues, post messages, or take actions in external services.
Also in 19 projects that use this
ArogyaReddySkillskill-share
A skill that creates new Claude skills and automatically shares them on Slack using Rube for seamless team collaboration and skill discovery.
Also in 17 projects that use this
ComposioHQSkillagent-deep-links
Build, validate, and troubleshoot deep links for Codex, Cursor, VS Code, Visual Studio, and similar tools. Use when users ask for clickable links (especially in Slack) that open threads, files, folders, or app settings.
Also in 3 projects that use this
ComposioHQSkillhelium-mcp
Search real-time news with bias scoring, get live stock/ETF/crypto data with AI analysis, ML options pricing, balanced news synthesis, and meme search via the Helium MCP server.
Also in 3 projects that use this
trendinuxSkilltwitter-algorithm-optimizer
Analyze and optimize tweets for maximum reach using Twitter's open-source algorithm insights. Rewrite and edit user tweets to improve engagement and visibility based on how the recommendation system ranks content.
Also in 12 projects that use this
LangChainSkilllangsmith-fetch
Fetches LangSmith traces for debugging agent behavior. Use when troubleshooting agent issues, reviewing conversation history, or investigating tool calls.
Also in 18 projects that use this
- +1 more from ComposioHQ
- +5 more from ArogyaReddy
Other versions64 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 64
- ComposioHQ/awesome-claude-skills★ 75ksame name
- ComposioHQ/awesome-codex-skills★ 17ksame name
- composio-community/awesome-codex-skills★ 16ksame name
- langchain-ai/open-swe★ 11ksame name
- suboss87/FDEOps★ 940same name
- indranilbanerjee/digital-marketing-pro★ 934same name
- ArogyaReddy/ComposioHQ-awesome-claude-skills★ 45same name
- frostant/awesome-claude-skills★ 44same name
- jexchan/dailyup-second-brain-starter★ 33same name
- microsoft/Employee-Self-Service-Agent-Developer-Kit★ 29same name
54 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository6 other files it gives its agents.
- adopt-projectSkill
- beginSkill
- next-lessonSkill
- plan-journeySkill
- start-projectSkill
- statusSkill
The file
We split this skill into the rules it gives an agent
12 rules. 0 are common, 4 are distinctive, and 8 are very specific to this project.
0
Common
Most of the 56 edited versions we read keep these.
4
Distinctive
This author's own choices. Worth a look.
8
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 16
Claude Code: run claude --version and take the leading semantic version number.
- Very specific: line 17
Codex: run codex --version and take the version from its codex-cli x.y.z output.
- Distinctive: line 18
If it is older than 0.131.0, warn the user that Altitude's session hooks need Codex CLI >= 0.131.0 and suggest updating first — the server …
- Very specific: line 21
GitHub Copilot CLI: run copilot --version and use the observed CLI version.
- Very specific: line 22
VS Code Copilot chat: read the installed GitHub Copilot Chat extension version from VS Code Extensions.
- Very specific: line 23
Cursor desktop: read the desktop version from cursor --version (first line) or Cursor About.
- Very specific: line 24
Cursor terminal: run cursor-agent --version (or agent --version if that is its installed command).
- Distinctive: line 31
If the status command fails or its output cannot be read, relay the short diagnostic and stop the pairing decision here.
- Distinctive: line 33
Start the pairing command for your agent in the background:
- Distinctive: line 41
Wait for the user to confirm in the browser, then report whether the command completed successfully.
- Very specific: line 42
Run altitude status and summarize the connection, binding, and journey state.
- Very specific: line 49
Do not alter the device flow or send credentials anywhere except through the altitude CLI.
“Common” is measured: kept by at least 6 in 10 of the 56 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Host commands: In Cursor desktop and Cursor terminal, invoke the same shared skills as /begin, /connect, /status, and /next-lesson (choose Altitude in the skill picker when names collide). Translate /altitude:<skill> examples in this file to /<skill> in Cursor; Claude Code keeps /altitude:<skill> and Codex keeps $<skill>.
Copilot host commands: In GitHub Copilot CLI and VS Code Copilot chat, use the same shared Altitude skills. Choose the Altitude skill by name in the host skill picker (connect, begin, next-lesson, or status); use the invocation spelling that host displays. Do not copy these files into a second Copilot-only skill library. The complete Altitude plugin supplies the lesson hooks; copied skills alone provide only the standalone free method.
Connect the current coding agent installation to Altitude:
Cursor preparation, including already-paired installations: before plugin setup or pairing, have the learner run altitude update in their own terminal and wait for it to finish. If altitude is missing, use npm install -g @learnaltitude/cli instead. Then read altitude --version from the same environment that will run pairing. Older CLIs do not recognize --agent cursor; a plugin update alone cannot add that identity. If the installed CLI still rejects Cursor after updating, keep the existing pairing/binding/progress and report that this CLI does not support Cursor yet; never re-run pairing without --agent or with another agent's identity. A successful update is not proof that a lesson passed acceptance.
- Identify which agent you are running in, then capture its version:
- Claude Code: run
claude --versionand take the leading semantic version number. - Codex: run
codex --versionand take the version from itscodex-cli x.y.zoutput. If it is older than 0.131.0, warn the user that Altitude's session hooks need Codex CLI >= 0.131.0 and suggest updating first — the server makes the final call during pairing. - GitHub Copilot CLI: run
copilot --versionand use the observed CLI version. - VS Code Copilot chat: read the installed GitHub Copilot Chat extension version from VS Code Extensions. Keep the editor version from
code --versionseparate; do not substitute it for the Copilot version. - Cursor desktop: read the desktop version from
cursor --version(first line) or Cursor About. Do not substitute the terminal agent version. - Cursor terminal: run
cursor-agent --version(oragent --versionif that is its installed command). Preserve its full date-and-build version string; do not invent a semantic version. - Cursor compatibility setup: resolve
../../bin/altitude-cursor-setup.mjsrelative to this installedskills/connect/SKILL.mdusing the source path attached to this invocation or the file you actually read. Do not assume plugin hook environment variables exist in your shell. If the source path is unavailable, recover the active installation location from Cursor's plugin manager; never choose the newest cache directory or another checkout. Then runnode '<that actual absolute file path>' install(quote for the current shell). This also records this exact installed plugin root as JSON data in~/.cursor/altitude/plugin.json; native hooks read that registration because Cursor can supply another plugin's root environment. Run/connectagain after every plugin update to refresh the registration from the newly active skill source. Missing or stale registrations fail open with a/connectdiagnostic. The setup idempotently adds three owned, inert user-hook entries that let the terminal agent dispatch the native plugin's turn hooks; existing hooks are preserved. Run the shipped script, never recreate its config by hand. If setup fails, relay its short error and leave the config intact. A copied skill without that script is not the full plugin: use Cursor's plugin manager to install the complete plugin. Start a new Cursor session after a changed setup before beginning a lesson. Removal uses the same installed script withuninstall; it removes only Altitude's compatibility entries and its matching registration; an older plugin cannot erase a newer registration. Cursor pairing decision: after compatibility setup, runaltitude statusbefore starting a new device flow. This command is local and cache-only:connectedreports whether a readable, correctly shaped saved token exists. It does not identify the account or verify that the server accepts the token; never call it proof of the intended account or healthy online authentication. - For ordinary initial setup or plugin-update maintenance with
connected: true, preserve the existing pairing and skip steps 2–4. Say the existing local pairing was kept and the plugin registration is ready; continue with the status summary and installation check below. A changed setup still needs a new Cursor session before/beginor/next-lesson. - If the learner explicitly asks to reconnect or change accounts, continue with pairing even when a local token exists. If a fresh CLI request has already returned
unauthorized, explain that the server rejected the saved credential and continue with pairing to repair it. Do not expose the credential or delete the binding, plan, or queued progress. - With
connected: false, the saved token is missing or malformed; continue with pairing. Offline or network-blocked reads, absent cached journey data, and an idle flusher do not imply an invalid token and never trigger automatic re-pairing. - If the status command fails or its output cannot be read, relay the short diagnostic and stop the pairing decision here; do not infer disconnection or start pairing merely because status failed. Preserve the existing local state and the registration just refreshed.
- Claude Code: run
Start the pairing command for your agent in the background:
- Claude Code:
altitude connect --agent claude-code --agent-version <version> --next-hint "Open your coding agent in your project folder and run /altitude:begin to start your journey." - Codex:
altitude connect --agent codex --agent-version <version> --next-hint "Open your coding agent in your project folder and run $begin to start your journey." - GitHub Copilot CLI or VS Code Copilot chat:
altitude connect --agent copilot --agent-version <observed Copilot version> --next-hint "Open your project in Copilot and choose Altitude begin to start your journey." - Cursor desktop or terminal:
altitude connect --agent cursor --agent-version <observed version> --next-hint "Open your coding agent in your project folder and run /begin to start your journey."
Codex aside — the sandbox: in Codex's default modes the shell has no internet access, so a plain background
altitude connectfails before it can print a code. Request escalated permissions for that one command with a one-line justification ("Altitude needs network access to pair this computer with your account"). If that is declined, or the host has no such mechanism, do not retry blindly and do not call the install broken: give the learner the exactaltitude connectline above to run in their own terminal, ask them to paste back the URL and code it prints, and continue from step 3 with those.- Claude Code:
- Relay the verification URL and one-time code from its output to the user immediately.
- Wait for the user to confirm in the browser, then report whether the command completed successfully. If the learner ran the command in their own terminal, ask them to tell you when it reports success instead.
- Run
altitude statusand summarize the connection, binding, and journey state. - Codex aside — the one-time trust prompt: Codex asks "Hooks need review" the first time it launches with Altitude installed, and again after a plugin update changes a hook. Tell the learner to choose Trust all and continue whenever it appears; if they dismissed it earlier, ask them to open
/hooksnow and trust Altitude's hooks. Those hooks run outside the sandbox and keep a local copy of the journey fresh, which is what lets$beginand$next-lessonread it from inside the sandbox; untrusted, every lesson starts by asking them to run a command in their own terminal. - Cursor installation check: the full native Altitude plugin must be loaded, including hooks; a copied
skills/folder serves only the standalone free method. Use Cursor Customize and its Hooks output to confirm loading, then start a new conversation. Preserve existing user and project hooks. This integration is awaiting full desktop/terminal lesson acceptance; do not claim a successful pairing proves the lesson works, or bypass a server compatibility pause.
Copilot installation check: install the complete native Altitude plugin, trust this project for hooks, and reopen the Copilot conversation in the intended project folder. The plugin must expose both the shared skills and its lifecycle hooks. In VS Code use the Local agent for the tested integration; other editor agent hosts require separate verification. Run altitude --version before pairing and check the CLI/server requirements in Copilot support. If --agent copilot is rejected, preserve existing pairing and progress and report the missing CLI support. Never omit --agent or pair as another agent to bypass compatibility. A fresh installation does not need an unconditional update; use the documented release-compatible version.
Do not alter the device flow or send credentials anywhere except through the altitude CLI.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

