Agent skill

Open Record Replay

by humblebanana in humblebanana/open-record-replay

Record a user-demonstrated macOS workflow with the Open Record/Replay CLI, inspect the captured event stream, prepare a skill evidence package, and hand it to the host agent's native skill creator.

MITAuto-check passedProductivity & Automation

Install Open Record Replay

skills CLI

$ npx skills add humblebanana/open-record-replay --skill open-record-replay -a claude-code

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

GitHub CLI

$ gh skill install humblebanana/open-record-replay open-record-replay --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/humblebanana/open-record-replay.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/open-record-replay .claude/skills/open-record-replay && 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/

Paste into Claude Code

Install the "open-record-replay" agent skill from https://github.com/humblebanana/open-record-replay/tree/main/skills/open-record-replay into .claude/skills/open-record-replay/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "open-record-replay", then confirm the skill loads.

Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

$skill-installer

$skill-installer install https://github.com/humblebanana/open-record-replay/tree/main/skills/open-record-replay

Type this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.

skills CLI

$ npx skills add humblebanana/open-record-replay --skill open-record-replay -a codex

Project install goes to .agents/skills/; add -g for ~/.codex/skills/.

GitHub CLI

$ gh skill install humblebanana/open-record-replay open-record-replay --agent codex

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/humblebanana/open-record-replay.git skills-src && mkdir -p .agents/skills && cp -r skills-src/skills/open-record-replay .agents/skills/open-record-replay && rm -rf skills-src

Use ~/.agents/skills/ instead of .agents/skills for a personal install.

Codex skills documentation · loads skills from .agents/skills/

Paste into Codex

Install the "open-record-replay" agent skill from https://github.com/humblebanana/open-record-replay/tree/main/skills/open-record-replay into .agents/skills/open-record-replay/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "open-record-replay", then confirm the skill loads.

Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

skills CLI

$ npx skills add humblebanana/open-record-replay --skill open-record-replay -a cursor

Project install goes to .agents/skills/; add -g for ~/.cursor/skills/.

GitHub CLI

$ gh skill install humblebanana/open-record-replay open-record-replay --agent cursor

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/humblebanana/open-record-replay.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/skills/open-record-replay .cursor/skills/open-record-replay && rm -rf skills-src

Use ~/.cursor/skills/ instead of .cursor/skills for a personal install.

Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/

Paste into Cursor

Install the "open-record-replay" agent skill from https://github.com/humblebanana/open-record-replay/tree/main/skills/open-record-replay into .cursor/skills/open-record-replay/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "open-record-replay", then confirm the skill loads.

Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

gemini skills install

$ gemini skills install https://github.com/humblebanana/open-record-replay.git --path skills/open-record-replay

--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.

skills CLI

$ npx skills add humblebanana/open-record-replay --skill open-record-replay -a gemini-cli

Project install goes to .agents/skills/; add -g for ~/.gemini/skills/.

GitHub CLI

$ gh skill install humblebanana/open-record-replay open-record-replay --agent gemini-cli

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/humblebanana/open-record-replay.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/skills/open-record-replay .gemini/skills/open-record-replay && rm -rf skills-src

Use ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.

Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/

Paste into Gemini CLI

Install the "open-record-replay" agent skill from https://github.com/humblebanana/open-record-replay/tree/main/skills/open-record-replay into .gemini/skills/open-record-replay/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "open-record-replay", then confirm the skill loads.

Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

GitHub CLI

$ gh skill install humblebanana/open-record-replay open-record-replay

Installs for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).

skills CLI

$ npx skills add humblebanana/open-record-replay --skill open-record-replay -a github-copilot

Project install goes to .agents/skills/; add -g for ~/.copilot/skills/.

Manual copy

$ git clone --depth 1 https://github.com/humblebanana/open-record-replay.git skills-src && mkdir -p .github/skills && cp -r skills-src/skills/open-record-replay .github/skills/open-record-replay && rm -rf skills-src

Use ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.

GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/

Paste into GitHub Copilot

Install the "open-record-replay" agent skill from https://github.com/humblebanana/open-record-replay/tree/main/skills/open-record-replay into .github/skills/open-record-replay/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "open-record-replay", then confirm the skill loads.

GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

skills CLI

$ npx skills add humblebanana/open-record-replay --skill open-record-replay -a opencode

OpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.

GitHub CLI

$ gh skill install humblebanana/open-record-replay open-record-replay --agent opencode

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/humblebanana/open-record-replay.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/skills/open-record-replay .opencode/skills/open-record-replay && rm -rf skills-src

Use ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.

OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/

Paste into OpenCode

Install the "open-record-replay" agent skill from https://github.com/humblebanana/open-record-replay/tree/main/skills/open-record-replay into .opencode/skills/open-record-replay/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "open-record-replay", then confirm the skill loads.

OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

Facts

Skill name: open-record-replay
GitHub stars: 144
Token cost: ~1.1k tokens
SKILL.md length: 483 words
Files: 1
Skills in repo: 1
Repo updated: Jun 22, 2026
First seen: Oct 7, 2026
Licence: MIT

At a glance

Record a user-demonstrated macOS workflow with the Open Record/Replay CLI, inspect the captured event stream, prepare a skill evidence package, and hand it to the host agent's native skill creator.

Works in 2 steps: Check permissions first → If permissions are missing, run
The user asks the agent to watch them perform a task
SKILL.md covers CLI Mapping, Recording Workflow, If The User Cancels and Interpreting Events, plus 2 more sections
Calls node

What it does

Open Record Replay is an agent skill from humblebanana/open-record-replay. Record a user-demonstrated macOS workflow with the Open Record/Replay CLI, inspect the captured event stream, prepare a skill evidence package, and hand it to the host agent's native skill creator. Use when the user asks the agent to watch them perform a task, record a workflow, or create/refine a reusable Computer Use skill from a demonstration.

Its SKILL.md is about 1.1k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Productivity & Automation, covering Desktop control and Skill authoring. It works with macOS. The repository describes itself as: Open-source macOS record-and-replay workflow recorder for computer use agents. Captures mouse, keyboard, and UI events as structured traces so agents can learn, replay, and… The licence is MIT.

When your agent uses it

The user asks the agent to watch them perform a task
Record a workflow
Create/refine a reusable Computer Use skill from a demonstration

Example prompts

“/open-record-replay”

Workflow steps

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

Check permissions first
If permissions are missing, run

What it can do on your machine

Read from SKILL.md and the folder at commit 8eff3fa. 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:
- node
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

Open Record Replay loads about 1.1k tokens when it runs. Until then it costs about 92 tokens; SKILL.md has 483 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it: ~92
When it runs · the whole SKILL.md, loaded when a task matches: ~1.1k

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 humblebanana/open-record-replay at commit 8eff3fa, republished under its MIT licence (© humblebanana). 483 words, ~1,110 tokens.

Download SKILL.mdSave it as .claude/skills/open-record-replay/SKILL.md (or your agent's skills folder).

name: open-record-replay
description: Record a user-demonstrated macOS workflow with the Open Record/Replay CLI, inspect the captured event stream, prepare a skill evidence package, and hand it to the host agent's native skill creator. Use when the user asks the agent to watch them perform a task, record a workflow, or create/refine a reusable Computer Use skill from a demonstration.

Open Record/Replay

Use this skill to learn a user-demonstrated macOS workflow through Open Record/Replay.

Open Record/Replay records evidence. It does not create the final skill by itself. After recording and packaging, invoke the current host agent's native Skill Creator or skill authoring flow.

CLI Mapping

The OpenAI Record & Replay tool names map to this project's CLI flow:

text

event_stream_start  -> node bin/orr.js record start
event_stream_stop   -> node bin/orr.js record stop
metadataPath        -> runs/sessions/<session-id>/session.json
eventsPath          -> runs/sessions/<session-id>/events.jsonl
skill handoff       -> node bin/orr.js skill prepare

Run commands from the Open Record/Replay repository root unless the user provides another path.

Recording Workflow

Check permissions first:

bash

node bin/orr.js permissions check

If permissions are missing, run:

bash

node bin/orr.js permissions request

Tell the user which macOS permissions are missing and wait for them to enable the permissions before recording.

Start recording only when the user is ready:

bash

node bin/orr.js record start --name my-workflow --out runs

After recording starts, do not poll or wait in a loop. End your turn and ask the user to perform the workflow and tell you when they are done.
When the user says they are done, stop the recording:

bash

node bin/orr.js record stop latest --out runs

Validate the recording:

bash

node bin/orr.js session validate-recording latest --out runs

Prepare a skill evidence package:

bash

node bin/orr.js skill prepare latest --runs runs --out skill-inputs

Confirm the package path and inspect the package files:

text

skill-inputs/<session-id>/
├── README.md
├── events.jsonl
└── session.json

If The User Cancels

If the user says they cancelled recording, do not continue to skill creation.

If needed, inspect session.json or orr_session.json to confirm cancellation, then acknowledge that no skill will be created from that recording.

Interpreting Events

Treat events.jsonl as the primary evidence. session.json gives timing and file paths only.

When inspecting events, pay attention to:

App and window attribution.
window.changed events.
mouse.click and mouse.drag targets.
keyboard.text_input and keyboard.submit targets.
selection.changed, selected text, and selected items.
AX full-tree or diff payloads.

Do not infer unsupported actions from generic targets such as AXGroup, AXScrollArea, or low-confidence action clusters. If the workflow is unclear, ask the user what the intended action was.

Show full SKILL.md (200 more words)Show less

Creating Or Refining A Skill

If the recording contains enough evidence to identify a reusable workflow, create or refine a real discoverable skill by using the current host agent's native Skill Creator.

Do not stop at a summary, replay plan, or Markdown runbook unless the user explicitly asks for only that.

When creating the final skill:

Treat the recording as evidence of the user's intended outcome, not a requirement to reproduce every raw event.
Prefer stable app/window/control targets from the event stream.
Include verification steps after side-effect actions such as send, upload, create, post, save, or publish.
Parameterize reusable inputs such as recipient, file path, URL, search query, prompt, or message text.
Avoid coordinate-only instructions unless no semantic target is available.
Ask the user when a key action or destination is ambiguous.

Privacy

Do not include sensitive values from recorded events in summaries or generated skills.

Treat the following as sensitive:

Passwords.
OTPs.
API keys or tokens.
SSNs, passports, financial account or card numbers.
Private personal, medical, legal, HR, or customer data.
Private local file paths or document names when not necessary for the workflow shape.

Use placeholders or generic descriptions when the skill needs to mention a sensitive value.

Files

Just SKILL.md in skills/open-record-replay of humblebanana/open-record-replay.

Open the folder on GitHubat commit 8eff3fa

Compare with similar skills

Open Record Replay 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.

Open Record Replay compared with similar skills
Skill	Stars	Used in	Tokens	Auto-check	Licence	Repo updated
Open Record Replay this skillhumblebanana/open-record-replay	144	—	~1.1k	Automated safety check: Pass	MIT	3 mo ago
Open Computer UseiFurySt/open-codex-computer-use	2.4k	—	~1.5k	Automated safety check: Pass	MIT	today
Mac Computer UseTo3akaRin/mac-computer-use	1.1k	—	~495	Automated safety check: Pass	MIT	17 days ago
Waku Computer Useegoist/waku	1.6k	—	~3.6k	Automated safety check: Pass	GPL-3.0	8 days ago
Interceptor BrowserHacker-Valley-Media/Interceptor	522	—	~4.8k	Automated safety check: Pass	Custom licence	8 days ago
Agent Desktopsuperagent-ai/grok-cli	3.5k	—	~720	Automated safety check: Pass	MIT	3 mo ago

Similar skills

Open Computer Use
iFurySt/open-codex-computer-use
Platform-neutral guidance for using Open Computer Use, the open-source Computer Use MCP server and CLI for macOS, Linux, and Windows.
2.4k GitHub stars~1.5k tokensUpdated today
Productivity & AutomationAuto-check passed
Mac Computer Use
To3akaRin/mac-computer-use
操作 macOS 桌面应用，探测窗口和自动化接口、截图、读取或修改辅助功能元素、执行鼠标键盘动作，以及通过 CDP 操作内嵌 Chromium 页面。适用于桌面应用自动化与界面验收；普通网页任务优先使用已有浏览器工具。
1.1k GitHub stars~495 tokensUpdated 17 days ago
Productivity & AutomationAuto-check passed
Waku Computer Use
egoist/waku
Control local macOS, Windows, and Linux apps through Waku Computer Use.
1.6k GitHub stars~3.6k tokensUpdated 8 days ago
Productivity & AutomationAuto-check passed
Interceptor Browser
Hacker-Valley-Media/Interceptor
Drive a signed-in Chrome / Brave / Safari session via the interceptor CLI: open/read pages, click, type, inspect DOM/text/network, automate rich browser editors and scene graphs, capture…
522 GitHub stars~4.8k tokensUpdated 8 days ago
Productivity & AutomationAuto-check passed
Agent Desktop
superagent-ai/grok-cli
Use the built-in Computer sub-agent with agent-desktop for macOS desktop automation.
3.5k GitHub stars~720 tokensUpdated 3 mo ago
Productivity & AutomationAuto-check passed
Interceptor iOS
Hacker-Valley-Media/Interceptor
Drive any installed app on an owned, unlocked, Developer-Mode iPhone via interceptor ios : ref-tagged element trees, deterministic coordinate taps (click), reliable text entry (type/keys), scroll…
522 GitHub stars~1.9k tokensUpdated 8 days ago
Productivity & AutomationAuto-check passed

Works with

macOS

Questions about Open Record Replay

What does Open Record Replay do?

Record a user-demonstrated macOS workflow with the Open Record/Replay CLI, inspect the captured event stream, prepare a skill evidence package, and hand it to the host agent's native skill creator. Open Record Replay is an agent skill from humblebanana/open-record-replay. Record a user-demonstrated macOS workflow with the Open Record/Replay CLI, inspect the captured event stream, prepare a skill evidence package, and hand it to the host agent's native skill creator.

When should I use Open Record Replay?

Open Record Replay fits situations like: the user asks the agent to watch them perform a task; record a workflow; create/refine a reusable Computer Use skill from a demonstration.

How do I install Open Record Replay in Claude Code?

Run `npx skills add humblebanana/open-record-replay --skill open-record-replay -a claude-code`. Or copy the skill folder (skills/open-record-replay in humblebanana/open-record-replay) into .claude/skills/open-record-replay in your project. Claude Code loads it when a task matches its description.

How do I install Open Record Replay in Codex?

Run `npx skills add humblebanana/open-record-replay --skill open-record-replay -a codex`. Or copy the skill folder (skills/open-record-replay in humblebanana/open-record-replay) into .agents/skills/open-record-replay in your project. Codex loads it when a task matches its description.

Can I use Open Record Replay 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 humblebanana/open-record-replay --skill open-record-replay -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/open-record-replay, .gemini/skills/open-record-replay, .github/skills/open-record-replay and .opencode/skills/open-record-replay in your project.

What does Open Record Replay need to run?

Going by SKILL.md and its folder, Open Record Replay needs the command-line tools its instructions call (node).

Does Open Record Replay 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 Open Record Replay 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 Open Record Replay use?

Open Record Replay 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 Open Record Replay use?

About 1.1k tokens (SKILL.md is roughly 4.4k 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 Open Record Replay?

Skills that share tags, products or a category with Open Record Replay: Open Computer Use (iFurySt/open-codex-computer-use, 2.4k stars), Mac Computer Use (To3akaRin/mac-computer-use, 1.1k stars), Waku Computer Use (egoist/waku, 1.6k stars) and Interceptor Browser (Hacker-Valley-Media/Interceptor, 522 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Open Record Replay?

humblebanana (a GitHub user) maintains it in humblebanana/open-record-replay, which has 144 GitHub stars. The repository was last updated on June 22, 2026.

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

Install Open Record Replay

Facts

At a glance

What it does

When your agent uses it

Example prompts

Workflow steps

What it can do on your machine

Tool permissions

Runs code

Network

Credentials

Context cost

Safety

SKILL.md

Open Record/Replay

CLI Mapping

Recording Workflow

If The User Cancels

Interpreting Events

Creating Or Refining A Skill

Privacy

Files

Compare with similar skills

Similar skills

Open Computer Use

Mac Computer Use

Waku Computer Use

Interceptor Browser

Agent Desktop

Interceptor iOS

Related

Questions about Open Record Replay