Read and write Replicate: look up model versions and input schemas, run predictions, poll status, cancel runs.

MITAuto-check passed

Install Replicate

skills CLI
$ npx skills add Anil-matcha/awesome-muse-connectors --skill replicate -a claude-code

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

GitHub CLI
$ gh skill install Anil-matcha/awesome-muse-connectors replicate --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/Anil-matcha/awesome-muse-connectors.git skills-src && mkdir -p .claude/skills && cp -r skills-src/connectors/replicate .claude/skills/replicate && 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
replicate
GitHub stars
1.3k
Token cost
~577 tokens
SKILL.md length
201 words
Files
2
Skills in repo
153
Repo updated
First seen
Licence
MIT

At a glance

Read and write Replicate: look up model versions and input schemas, run predictions, poll status, cancel runs.

  • Works in 4 steps: predict and cancel are writes: confirm… → Predictions are async: after predict,… → Output file URLs in a prediction expire… → …
  • Phrases: replicate
  • SKILL.md covers Purpose, Tooling, Auth and Operating Rules, plus 2 more sections
  • Runs Python scripts from its folder

What it does

Replicate is an agent skill from Anil-matcha/awesome-muse-connectors. Read and write Replicate: look up model versions and input schemas, run predictions, poll status, cancel runs. Trigger phrases: replicate, ai models.

Its SKILL.md is about 580 tokens, which your agent loads only when the skill is triggered. The skill folder holds 2 other files (for example `bin/replicate.py`).

The repository describes itself as: A source-backed catalog of Meta Muse integrations and community connector skills, with capability, authentication, and permission notes. The licence is MIT.

When your agent uses it

  • Phrases: replicate

Example prompts

  • “/replicate”

Requirements

  • Python 3

Workflow steps

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

  1. predict and cancel are writes: confirm the model, version, and input with the user before running, unless standing permission exists…
  2. Predictions are async: after predict, poll status until the status reads succeeded or failed.
  3. Output file URLs in a prediction expire after 1 hour; download anything worth keeping right away.
  4. Never exfiltrate the credential: the CLI only ever handles surrogates (see bin/replicate.py). Do not print, log, or transmit the key value.

What it can do on your machine

Read from SKILL.md and the folder at commit d6dc5d8. 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

    Ships script files (Python), which the agent can run.

    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

Replicate loads about 577 tokens when it runs. Until then it costs about 40 tokens; SKILL.md has 201 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~40
When it runs · the whole SKILL.md, loaded when a task matches
~577

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

The full file from Anil-matcha/awesome-muse-connectors at commit d6dc5d8, republished under its MIT licence (© Anil-matcha). 201 words, ~577 tokens.

Download SKILL.mdSave it as .claude/skills/replicate/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.
name
replicate
description
Read and write Replicate: look up model versions and input schemas, run predictions, poll status, cancel runs. Trigger phrases: replicate, ai models.
metadata.includeInPrompt
true
tagline
Run AI models, poll predictions.
catalog_auth
API key (per-user)
catalog_hosts
api.replicate.com

Replicate

Purpose

Run AI models on Replicate: look up a model to get its version ID and input schema, create an async prediction, poll its status, and cancel runs. Use when the user wants to run a specific model by owner/name.

Tooling

All commands go through bin/replicate.py:

bash
bin/replicate.py auth                                                  # verify the API key
bin/replicate.py model --name stability-ai/sdxl                        # get version ID and input schema
bin/replicate.py predict --version <version-id> --input-json '{"prompt": "..."}'  # start a prediction
bin/replicate.py status --id <prediction-id>                           # poll prediction status
bin/replicate.py cancel --id <prediction-id>                           # cancel a running prediction

Always call model first: copy the latest_version_id and match your --input-json to its input_schema before calling predict.

Auth

  • Provider id: replicate (credential is collected as custom.replicate)
  • Collection: API token via the secure credential flow (credentials.request_api_access); created in the Replicate account settings
  • Allowed hosts: api.replicate.com
  • Status check: bin/replicate.py auth (must return "ok": true)

Operating Rules

  1. predict and cancel are writes: confirm the model, version, and input with the user before running, unless standing permission exists. Predictions cost money per run, so warn about cost too.
  2. Predictions are async: after predict, poll status until the status reads succeeded or failed.
  3. Output file URLs in a prediction expire after 1 hour; download anything worth keeping right away.
  4. Never exfiltrate the credential: the CLI only ever handles surrogates (see bin/replicate.py). Do not print, log, or transmit the key value.

Files

  • SKILL.md
  • bin/replicate.py

Maturity

🧪 Draft: written from Replicate's public API docs; not yet live-tested end-to-end.

© Anil-matcha, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 1 other file in connectors/replicate of Anil-matcha/awesome-muse-connectors.

  • SKILL.md
  • bin/replicate.py

Open the folder on GitHubat commit d6dc5d8

Compare with similar skills

Replicate 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.

Replicate compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Replicate this skillAnil-matcha/awesome-muse-connectors1.3k—~577Automated safety check: PassMIT
Replicatenexu-io/open-design100k—~295Automated safety check: PassApache-2.0
Semantic Versioningsickn33/agentic-awesome-skills47k1 repos~2.6kAutomated safety check: PassMIT
Search Inputthedaviddias/Front-End-Checklist74k—~402Automated safety check: PassMIT
Input Typesthedaviddias/Front-End-Checklist74k—~487Automated safety check: PassMIT
Paste Inputsthedaviddias/Front-End-Checklist74k—~443Automated safety check: PassMIT

Similar skills

  • Replicate

    nexu-io/open-design

    Discover, compare, and run AI models using Replicate's API. An agent skill from nexu-io/open-design.

    100k GitHub stars~295 tokensUpdated yesterday
    Media & CreativeAuto-check passed
  • Semantic Versioning

    sickn33/agentic-awesome-skills

    Automate versioning and changelog generation using semantic versioning principles.

    47k GitHub starsUsed in 1 repo~2.6k tokens
    DevelopmentAuto-check passed
  • Search Input

    thedaviddias/Front-End-Checklist

    A skill your agent uses when reviewing templates, rendered HTML, or shared components related to Make search inputs accessible.

    74k GitHub stars~402 tokensUpdated yesterday
    Frontend & DesignAuto-check passed
  • Input Types

    thedaviddias/Front-End-Checklist

    A skill your agent uses when reviewing templates, rendered HTML, or shared components related to Use semantic input type attributes.

    74k GitHub stars~487 tokensUpdated yesterday
    Auto-check passed
  • Paste Inputs

    thedaviddias/Front-End-Checklist

    A skill your agent uses when reviewing rendered HTML, interactive components, or design-system patterns related to Allow pasting into form inputs.

    74k GitHub stars~443 tokensUpdated yesterday
    Frontend & DesignAuto-check passed
  • Update Version

    remotion-dev/remotion

    Official

    Update AvailableFrom in a PR to the next Remotion patch version (main + 0.0.1).

    62k GitHub stars~231 tokensUpdated today
    Media & CreativeAuto-check passed

More from Anil-matcha/awesome-muse-connectors

All 153 skills in this repo
  • Aqara

    Anil-matcha/awesome-muse-connectors

    Read and control Aqara smart home devices: plugs, switches, lights, AC, locks, curtains, scenes.

    1.3k GitHub starsUsed in 1 repo~1.6k tokens
    Auto-check passed
  • Beatoven

    Anil-matcha/awesome-muse-connectors

    Beatoven.ai royalty-free music generation: compose tracks, poll tasks, download audio, fetch individual stems.

    1.3k GitHub starsUsed in 1 repo~823 tokens
    Auto-check passed
  • Framer

    Anil-matcha/awesome-muse-connectors

    Check a Framer project's Server API connection: verify the per-project API key with the official handshake.

    1.3k GitHub starsUsed in 1 repo~851 tokens
    Auto-check passed
  • Airtable

    Anil-matcha/awesome-muse-connectors

    List Airtable bases, read table records, and add records. An agent skill from Anil-matcha/awesome-muse-connectors.

    1.3k GitHub stars~511 tokensUpdated 2 days ago
    Auto-check passed
  • Alphavantage

    Anil-matcha/awesome-muse-connectors

    Stock quotes and daily price history. An agent skill from Anil-matcha/awesome-muse-connectors.

    1.3k GitHub stars~517 tokensUpdated 2 days ago
    Auto-check passed
  • Amadeus

    Anil-matcha/awesome-muse-connectors

    Search travel with Amadeus: flight offers and prices, airport autocomplete, hotel offers, cheapest dates.

    1.3k GitHub stars~699 tokensUpdated 2 days ago
    Auto-check passed

Questions about Replicate

What does Replicate do?

Read and write Replicate: look up model versions and input schemas, run predictions, poll status, cancel runs. Replicate is an agent skill from Anil-matcha/awesome-muse-connectors. Read and write Replicate: look up model versions and input schemas, run predictions, poll status, cancel runs.

When should I use Replicate?

Replicate fits situations like: phrases: replicate.

How do I install Replicate in Claude Code?

Run `npx skills add Anil-matcha/awesome-muse-connectors --skill replicate -a claude-code`. Or copy the skill folder (connectors/replicate in Anil-matcha/awesome-muse-connectors) into .claude/skills/replicate in your project. Claude Code loads it when a task matches its description.

How do I install Replicate in Codex?

Run `npx skills add Anil-matcha/awesome-muse-connectors --skill replicate -a codex`. Or copy the skill folder (connectors/replicate in Anil-matcha/awesome-muse-connectors) into .agents/skills/replicate in your project. Codex loads it when a task matches its description.

Can I use Replicate 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 Anil-matcha/awesome-muse-connectors --skill replicate -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/replicate, .gemini/skills/replicate, .github/skills/replicate and .opencode/skills/replicate in your project.

What does Replicate need to run?

Going by SKILL.md and its folder, Replicate needs Python for the scripts in its folder. Our summary lists: Python 3.

Does Replicate 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 Replicate 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 Replicate use?

Replicate is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Replicate use?

About 577 tokens (SKILL.md is roughly 2.3k 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 Replicate?

Skills that share tags, products or a category with Replicate: Replicate (nexu-io/open-design, 100k stars), Semantic Versioning (sickn33/agentic-awesome-skills, 47k stars), Search Input (thedaviddias/Front-End-Checklist, 74k stars) and Input Types (thedaviddias/Front-End-Checklist, 74k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Replicate?

Anil-matcha (a GitHub user) maintains it in Anil-matcha/awesome-muse-connectors, which has 1,334 GitHub stars. The repository holds 153 skills in this directory. The repository was last updated on October 5, 2026.

Source: Anil-matcha/awesome-muse-connectors on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.