Agent skill

Refly Skill Runner

by refly-ai in refly-ai/refly

Base skill for the Refly ecosystem: finds, runs and monitors workflow-backed skills through the refly command line, with the actual work done on the Refly backend.

Custom licenceAuto-check passedProductivity & Automation

Install Refly Skill Runner

skills CLI
$ npx skills add refly-ai/refly --skill refly -a claude-code

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

GitHub CLI
$ gh skill install refly-ai/refly refly --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/refly-ai/refly.git skills-src && mkdir -p .claude/skills && cp -r skills-src/packages/cli/skill .claude/skills/refly && 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
refly
GitHub stars
7.5k
Token cost
~1.7k tokens
SKILL.md length
448 words
Files
1
Skills in repo
1
Repo updated
First seen
Licence
Custom licence

At a glance

Base skill for the Refly ecosystem: finds, runs and monitors workflow-backed skills through the refly command line, with the actual work done on the Refly backend.

  • Works in 5 steps: CLI only - Use refly , never call API… → Trust JSON - Only trust CLI JSON (ok,… → No fabricated IDs - Never invent… → …
  • Running an installed Refly skill and collecting its output files
  • SKILL.md covers Rules, Available Commands, Skill Categories & Execution… and Directory Structure, plus 2 more sections
  • Calls jq

What it does

Refly creates, discovers and runs domain skills that are bound to workflows, and it hands execution to the Refly backend instead of doing the work locally. Everything goes through the `refly` command line. The agent is told to trust only the CLI's JSON output, never invent workflow, run or node IDs, never print or request auth tokens, and to stop and show the hint when a call reports an error.

The command table covers `refly status`, `refly login`, listing marketplace skills and your installed ones, `refly skill run` with an installation ID and JSON input, watching a run with `--watch` (polling every 5000 ms by default), reading run details and tool calls, and downloading result files. Execution follows a pattern by output type: image, video and audio skills are run, awaited, downloaded and opened; search and report skills return text; email and messaging skills return a status to confirm. The excerpt ends inside the first pattern.

When your agent uses it

  • Running an installed Refly skill and collecting its output files
  • Checking the progress or result of a Refly workflow run
  • Listing the skills available in the Refly marketplace

Example prompts

  • “List my installed Refly skills and run the image generator with a prompt about a lighthouse at dusk.”
  • “Watch the latest Refly workflow run and download the files it produced.”
  • “Check whether I am logged in to Refly.”

Requirements

  • The @refly/cli package installed and authenticated

Workflow steps

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

  1. CLI only - Use refly , never call API directly.
  2. Trust JSON - Only trust CLI JSON (ok, payload, error, hint).
  3. No fabricated IDs - Never invent workflow/run/node IDs.
  4. No tokens - Never print or request auth tokens.
  5. Stop on error - If ok=false, stop and show hint.

What it can do on your machine

Read from SKILL.md and the folder at commit 71f8b87. 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:

    • jq

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

  • Network

    No URLs in SKILL.md.

    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

Refly Skill Runner loads about 1.7k tokens when it runs. Until then it costs about 110 tokens; SKILL.md has 448 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~110
When it runs · the whole SKILL.md, loaded when a task matches
~1.7k

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 448 words (~1,655 tokens).

“Tip: Get installationId (skpi-xxx) from refly skill installations.”

— opening of SKILL.md by refly-ai, Custom licence
name
refly

Read the full SKILL.md on GitHub

Files

Just SKILL.md in packages/cli/skill of refly-ai/refly.

Open the folder on GitHubat commit 71f8b87

Compare with similar skills

Refly Skill Runner 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.

Refly Skill Runner compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Refly Skill Runner this skillrefly-ai/refly7.5k—~1.7kAutomated safety check: PassCustom licence
Piece Builderactivepieces/activepieces25k—~3.1kAutomated safety check: NotesCustom licence
Profile App Cpuactivepieces/activepieces25k—~1.9kAutomated safety check: PassCustom licence
Profile Worker Memoryactivepieces/activepieces25k—~1.7kAutomated safety check: PassCustom licence
Build Audit Logsactivepieces/activepieces25k1 repos~5.9kAutomated safety check: PassCustom licence
Aeon Agent Setup and Managementaeonfun/aeon767—~9kAutomated safety check: WarnMIT

Similar skills

  • Piece Builder

    activepieces/activepieces

    Build and edit Activepieces pieces (integrations) — creating new pieces, adding actions or triggers, or fixing bugs in existing ones.

    25k GitHub stars~3.1k tokensUpdated today
    Productivity & AutomationAuto-check: notes
  • Profile App Cpu

    activepieces/activepieces

    Find what is burning CPU in a live app container without restarting it: separate the JS thread from GC and libuv, open the V8 inspector in place with SIGUSR1, take a CPU profile over CDP, and…

    25k GitHub stars~1.9k tokensUpdated today
    Productivity & AutomationAuto-check passed
  • Profile Worker Memory

    activepieces/activepieces

    Diagnose a worker that is growing memory or being OOM-killed: separate a JS-heap leak from native growth, take a V8 heap snapshot of a live worker in place (never uploading it), and walk the…

    25k GitHub stars~1.7k tokensUpdated today
    Productivity & AutomationAuto-check passed
  • Build Audit Logs

    activepieces/activepieces

    Build or review audit trails in TypeScript/JavaScript apps using evlog (pipelines, typed actions, denials, retention, compliance-style reviews).

    25k GitHub starsUsed in 1 repo~5.9k tokens
    Productivity & AutomationAuto-check passed
  • Sets up and manages an Aeon agent instance that runs skills on a schedule through GitHub Actions: starting, rescheduling, debugging, editing skills and mining chat history.

    767 GitHub stars~9k tokensUpdated today
    Productivity & AutomationAuto-check: warnings
  • Fastgpt Workflow Generator

    YYH211/Claude-meta-skill

    Generates production-ready FastGPT workflow JSON from natural language requirements.

    282 GitHub stars~5.5k tokensUpdated 4 mo ago
    Productivity & AutomationAuto-check passed

Questions about Refly Skill Runner

What does Refly Skill Runner do?

Base skill for the Refly ecosystem: finds, runs and monitors workflow-backed skills through the refly command line, with the actual work done on the Refly backend. Refly creates, discovers and runs domain skills that are bound to workflows, and it hands execution to the Refly backend instead of doing the work locally. Everything goes through the `refly` command line.

When should I use Refly Skill Runner?

Refly Skill Runner fits situations like: running an installed Refly skill and collecting its output files; checking the progress or result of a Refly workflow run; listing the skills available in the Refly marketplace.

How do I install Refly Skill Runner in Claude Code?

Run `npx skills add refly-ai/refly --skill refly -a claude-code`. Or copy the skill folder (packages/cli/skill in refly-ai/refly) into .claude/skills/refly in your project. Claude Code loads it when a task matches its description.

How do I install Refly Skill Runner in Codex?

Run `npx skills add refly-ai/refly --skill refly -a codex`. Or copy the skill folder (packages/cli/skill in refly-ai/refly) into .agents/skills/refly in your project. Codex loads it when a task matches its description.

Can I use Refly Skill Runner 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 refly-ai/refly --skill refly -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/refly, .gemini/skills/refly, .github/skills/refly and .opencode/skills/refly in your project.

What does Refly Skill Runner need to run?

Going by SKILL.md and its folder, Refly Skill Runner needs the command-line tools its instructions call (jq). Our summary lists: The @refly/cli package installed and authenticated.

Does Refly Skill Runner access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Refly Skill Runner 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 Refly Skill Runner use?

Refly Skill Runner 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 Refly Skill Runner use?

About 1.7k tokens (SKILL.md is roughly 6.6k 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 Refly Skill Runner?

Skills that share tags, products or a category with Refly Skill Runner: Piece Builder (activepieces/activepieces, 25k stars), Profile App Cpu (activepieces/activepieces, 25k stars), Profile Worker Memory (activepieces/activepieces, 25k stars) and Build Audit Logs (activepieces/activepieces, 25k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Refly Skill Runner?

refly-ai (a GitHub organization) maintains it in refly-ai/refly, which has 7,532 GitHub stars. The repository was last updated on July 29, 2026.

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