Agent skill

Refine Live In-Chat Relay

by Jakubantalik in Jakubantalik/transitions.dev

Acts as the LLM behind the Timeline Inspector's Refine button by long-polling a local relay for jobs, posting suggestions back, and exiting cleanly on a stop signal.

Custom licenceAuto-check passedFrontend & Design

Install Refine Live In-Chat Relay

skills CLI
$ npx skills add Jakubantalik/transitions.dev --skill refine-live -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install Jakubantalik/transitions.dev refine-live --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/Jakubantalik/transitions.dev.git skills-src && mkdir -p .claude/skills && cp -r skills-src/refine/.agents/skills/refine-live .claude/skills/refine-live && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
refine-live
GitHub stars
4.7k
Token cost
~6k tokens
SKILL.md length
2,579 words
Files
1
Skills in repo
3
Repo updated
First seen
Licence
Custom licence

At a glance

Acts as the LLM behind the Timeline Inspector's Refine button by long-polling a local relay for jobs, posting suggestions back, and exiting cleanly on a stop signal.

  • Works in 3 steps: Relay stop signal (authoritative). GET… → The user says so — "stop refine", "exit… → Your own idle backoff (safety net). A…
  • Running the Refine button's LLM tab without a persistent agent CLI wired up
  • SKILL.md covers Two modes, How it works, The loop — stay live, but… and Scan jobs (group from source), plus 3 more sections
  • Calls curl and npx

What it does

The skill is explicitly a fallback: a persistent CLI mode is preferred because it spawns an agent per click with no idle cost, and this in-chat loop should only be used when that persistent agent isn't wired up. Once started, the agent becomes the poller against a local relay, typically at localhost:7331, repeatedly calling an endpoint for the next job; each click in the browser's Refine panel sends one transition job, which the agent reasons about and posts a result back for.

Three exits are defined in priority order: an authoritative stop signal from the relay (sent when the user clicks Stop in the panel, or automatically after about ten minutes of no jobs) must always end the loop immediately; the user saying something like stop refine; and the agent's own idle backoff, which re-polls quickly at first and then backs off as empty responses accumulate, so a forgotten loop doesn't keep burning chat turns. The agent never re-polls after a stop signal.

When your agent uses it

  • Running the Refine button's LLM tab without a persistent agent CLI wired up
  • Answering transition refine jobs sent from the Timeline Inspector panel
  • Exiting an in-chat refine loop cleanly on a stop signal

Example prompts

  • “/refine live, please start answering jobs”
  • “Go live and start answering refine jobs from the relay.”
  • “Stop refine now, I'm done for today.”

Requirements

  • The transitions-refine relay running locally with no REFINE_AGENT_CMD wired

Workflow steps

3 steps, taken from the first numbered list in SKILL.md.

  1. Relay stop signal (authoritative). GET /jobs/next may return 200 with
  2. The user says so — "stop refine", "exit live", etc.
  3. Your own idle backoff (safety net). A long stretch of 204s is normal —

What it can do on your machine

Read from SKILL.md and the folder at commit 2773886. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • curl
    • npx

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md. Its commands use curl and npx, which can reach the network depending on how they are called.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Refine Live In-Chat Relay loads about 6k tokens when it runs. Until then it costs about 108 tokens; SKILL.md has 2,579 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~108
When it runs · the whole SKILL.md, loaded when a task matches
~6k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

Its licence (Custom licence) doesn't allow us to republish the file, so here is its outline and opening line. It has 2,579 words (~5,962 tokens).

“Persistent (recommended — run and forget) Run npx transitions-refine live from your project. The CLI starts the relay and wires REFINE_AGENT_CMD so the relay spawns your agent CLI per Refine click. No chat loop; idle = zero credit burn. Works…”

— opening of SKILL.md by Jakubantalik, Custom licence
name
refine-live

Read the full SKILL.md on GitHub

Files

Just SKILL.md in refine/.agents/skills/refine-live of Jakubantalik/transitions.dev.

Open the folder on GitHubat commit 2773886

Compare with similar skills

Refine Live In-Chat Relay next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Refine Live In-Chat Relay compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Refine Live In-Chat Relay this skillJakubantalik/transitions.dev4.7k—~6kAutomated safety check: PassCustom licence
Impeccablebestofjs/bestofjs3.1k26 repos~2.6kAutomated safety check: PassMIT
GSAP ScrollTriggergreensock/gsap-skills16k4 repos~4.6kAutomated safety check: PassMIT
Fixing Motion Performanceibelick/ui-skills9.6k5 repos~1.4kAutomated safety check: PassMIT
GSAP Core Animationgreensock/gsap-skills16k3 repos~3.7kAutomated safety check: PassMIT
Web Animation DesignbaptisteArno/typebot.io11k2 repos~2.7kAutomated safety check: PassCustom licence

Similar skills

  • Impeccable

    bestofjs/bestofjs

    A skill your agent uses when the user wants to design, redesign, shape, critique, audit, polish, clarify, distill, harden, optimize, adapt, animate, colorize, extract, or otherwise improve a…

    3.1k GitHub starsUsed in 26 repos~2.6k tokens
    Frontend & DesignAuto-check passed
  • GSAP ScrollTrigger

    greensock/gsap-skills

    Teaches an agent how to wire up GSAP's ScrollTrigger plugin for scroll-linked animation, pinned sections and scrub, including the start and end position syntax.

    16k GitHub starsUsed in 4 repos~4.6k tokens
    Frontend & DesignAuto-check passed
  • Fixing Motion Performance

    ibelick/ui-skills

    Audits and fixes web animation performance: layout thrashing, work that belongs on the compositor, scroll-linked motion and costly blur effects.

    9.6k GitHub starsUsed in 5 repos~1.4k tokens
    Frontend & DesignAuto-check passed
  • GSAP Core Animation

    greensock/gsap-skills

    Covers the GSAP core API for tweens, easing, staggers, defaults and matchMedia, and when to choose GSAP over CSS animations or other JavaScript animation libraries.

    16k GitHub starsUsed in 3 repos~3.7k tokens
    Frontend & DesignAuto-check passed
  • Web Animation Design

    baptisteArno/typebot.io

    Guides easing, timing and animation choices for UI motion, based on a web animation course, and reviews existing animations in a before-and-after table.

    11k GitHub starsUsed in 2 repos~2.7k tokens
    Frontend & DesignAuto-check passed
  • Scroll World Landing Page

    oso95/scroll-world

    Builds a scroll-driven landing page where a pre-rendered camera flies through connected AI-generated scenes, using Higgsfield for stills and video clips.

    9.9k GitHub starsUsed in 1 repo~12k tokens
    Frontend & DesignAuto-check: notes

More from Jakubantalik/transitions.dev

  • Transitions Polish

    Jakubantalik/transitions.dev

    Audits existing UI animations against the transitions.dev motion-token scale and suggests tokens for duration, distance, scale, blur and easing.

    4.7k GitHub starsUsed in 2 repos~3.2k tokens
    Auto-check passed
  • Transitions Agent Motion Audit

    Jakubantalik/transitions.dev

    Scans a codebase for missing or inconsistent UI motion, scores it from 0 to 100, and applies a transitions.dev fix only after you pick between a light polish and a full recipe revamp.

    4.7k GitHub stars~1.6k tokensUpdated today
    Auto-check passed

Questions about Refine Live In-Chat Relay

What does Refine Live In-Chat Relay do?

Acts as the LLM behind the Timeline Inspector's Refine button by long-polling a local relay for jobs, posting suggestions back, and exiting cleanly on a stop signal. The skill is explicitly a fallback: a persistent CLI mode is preferred because it spawns an agent per click with no idle cost, and this in-chat loop should only be used when that persistent agent isn't wired up. Once started, the agent becomes the poller against a local relay, typically at localhost:7331, repeatedly calling an endpoint for the next job; each click in the browser's Refine panel sends one transition job, which the agent reasons about and posts a result back for.

When should I use Refine Live In-Chat Relay?

Refine Live In-Chat Relay fits situations like: running the Refine button's LLM tab without a persistent agent CLI wired up; answering transition refine jobs sent from the Timeline Inspector panel; exiting an in-chat refine loop cleanly on a stop signal.

How do I install Refine Live In-Chat Relay in Claude Code?

Run `npx skills add Jakubantalik/transitions.dev --skill refine-live -a claude-code`. Or copy the skill folder (refine/.agents/skills/refine-live in Jakubantalik/transitions.dev) into .claude/skills/refine-live in your project. Claude Code loads it when a task matches its description.

How do I install Refine Live In-Chat Relay in Codex?

Run `npx skills add Jakubantalik/transitions.dev --skill refine-live -a codex`. Or copy the skill folder (refine/.agents/skills/refine-live in Jakubantalik/transitions.dev) into .agents/skills/refine-live in your project. Codex loads it when a task matches its description.

Can I use Refine Live In-Chat Relay in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add Jakubantalik/transitions.dev --skill refine-live -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/refine-live, .gemini/skills/refine-live, .github/skills/refine-live and .opencode/skills/refine-live in your project.

What does Refine Live In-Chat Relay need to run?

Going by SKILL.md and its folder, Refine Live In-Chat Relay needs the command-line tools its instructions call (curl and npx). Our summary lists: The transitions-refine relay running locally with no REFINE_AGENT_CMD wired.

Does Refine Live In-Chat Relay access the network?

SKILL.md contains no URLs. Its commands use curl and npx, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is Refine Live In-Chat Relay safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Refine Live In-Chat Relay use?

Refine Live In-Chat Relay has a licence file (the repository's licence) that doesn't match a standard licence. Read it on GitHub before reusing the skill.

How many tokens does Refine Live In-Chat Relay use?

About 6k tokens (SKILL.md is roughly 24k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Refine Live In-Chat Relay?

Skills that share tags, products or a category with Refine Live In-Chat Relay: Impeccable (bestofjs/bestofjs, 3.1k stars), GSAP ScrollTrigger (greensock/gsap-skills, 16k stars), Fixing Motion Performance (ibelick/ui-skills, 9.6k stars) and GSAP Core Animation (greensock/gsap-skills, 16k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Refine Live In-Chat Relay?

Jakubantalik (a GitHub user) maintains it in Jakubantalik/transitions.dev, which has 4,747 GitHub stars. The repository holds 3 skills in this directory. The repository was last updated on October 10, 2026.

Source: Jakubantalik/transitions.dev on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.