agentsclimarketplace

Chatgpt

Skill mediause/agent-skills/chatgpt

MediaUse Skills, web operations into semantic CLI actions, AI agents can run with near-zero token cost and higher stability.

Install
npx -y skills add mediause/agent-skills --skill chatgpt

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
  • 9 stars9 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Standardized ChatGPT skill for MediaUse. Includes Windows install, key onboarding, strict context/auth flow, full ChatGPT dynamic command map, and safety/rate controls.

SKILL.md

8.7 KB, as published. Nobody here has run it

MediaUse ChatGPT Skill

Use this skill to run structured ChatGPT workflows in MediaUse, including conversation control, prompt/response retrieval, and image generation with safe execution pacing.

Scope

Use this skill when the task targets ChatGPT operations such as:

  • Account: health check
  • Chat: status, new conversation, send/ask prompts, read/history/detail
  • Image: generate images from prompts

1. Install MediaUse CLI (Windows Only)

Use the official install script for Windows:

Run:

powershell -C "iwr https://release.mediause.dev/install.ps1 -UseBasicParsing | iex"

Then verify :

mediause --version

Current support status:

  • Windows: supported
  • Linux: not supported yet
  • macOS: not supported yet

Recommended skill install path:

  • .mediause/skills/chatgpt/SKILL.md

2. Get and Configure MediaUse Key

2.1 Apply for key

  1. Open https://mediause.dev/
  2. Sign in to your account.
  3. Open Project.
  4. Create or copy your API key.

2.2 Configure key in CLI

mediause manage key <your_key> --json

3. Core Flow (Mandatory Order)

Always follow this order:

First step on every run: execute install script once to auto-install/auto-upgrade the latest MediaUse CLI.

  1. Discover site and commands.
  2. Bind account context with use account.
  3. Check status with auth health.
  4. Execute dynamic site actions.
  5. Verify with trace/task.

3.1 Discover and plugin setup

mediause plugin list --json
mediause plugin add chatgpt --json
mediause chatgpt -h
mediause chatgpt chat -h

3.2 Bind context before any read/write

use account must be executed successfully before any fetch or interaction action.

use account argument format:

  • <platform:account_id>
  • account_id should be selected from mediause auth list --json.

By default, use account keeps the browser hidden. Use --show when manual observation or takeover is needed.

mediause auth list --json
mediause use account chatgpt:<account_id> --policy balanced --json

3.3 Auth health precondition

auth health is valid only after successful use account.

mediause auth health --json

If auth health indicates not logged in/expired:

mediause auth login chatgpt --json
mediause use account chatgpt:<account_id> --policy balanced --json
mediause auth health --json

If page shows unusual traffic, captcha, or risk confirmation:

mediause use account chatgpt:<account_id> --policy balanced --show --json

Risk reduction note:

  • In challenged sessions, --show usually reduces repeated interception risk.

Complete verification manually, then rerun the action.

3.4 Guest mode (optional)

Manifest default account id is guest. Guest mode can be used only for read/chat actions when the runtime supports unauthenticated chatgpt:guest context (that is, mediause use account chatgpt:guest --json succeeds):

mediause use account chatgpt:guest --json

Guest mode rules:

  • Read/chat operations only (chat.status, chat.new, chat.send, chat.ask, chat.read, chat.history, chat.detail).
  • image.generate may require logged-in account depending on current runtime and page state.
  • If action is blocked in guest mode, switch to logged-in account context.

4. ChatGPT Dynamic Command Map (v1)

Source schema:

  • plugin: plugin.chatgpt
  • manifest: crates/platforms/plugins/chatgpt/manifest.yaml

4.1 account.*

  • mediause chatgpt account health --json

4.2 chat.*

  • mediause chatgpt chat status --json
  • mediause chatgpt chat new --json
  • mediause chatgpt chat send --prompt <text> [--new <bool>] --json
  • mediause chatgpt chat ask --prompt <text> [--timeout <seconds>] [--new <bool>] --json
  • mediause chatgpt chat read [--markdown <bool>] --json
  • mediause chatgpt chat history [--limit <n>] --json
  • mediause chatgpt chat detail --id <conversation_id> [--markdown <bool>] --json

4.3 image.*

  • mediause chatgpt image generate --prompt <text> [--timeout <seconds>] --json

5. Operational Constraints (Mandatory)

Apply these constraints for all actions to reduce account risk and keep behavior stable.

5.1 Human-like pacing

  • Never execute high-frequency prompt bursts continuously.
  • Add randomized delay between repeated prompt/image requests.
  • Add longer cooldown after long-running ask/generate operations.

5.2 Frequency limits and minimum spacing

  • Hard stop if operation rate is abnormally high.
  • Stop immediately on repeated anti-bot challenge, login re-validation, or risk prompt.

Quick policy matrix:

  • chat.ask / chat.send: <= 30 per hour, >= 5s spacing
  • image.generate: <= 10 per hour, >= 30s spacing
  • chat.read / chat.history / chat.detail / chat.status: <= 60 per minute, >= 1s spacing

If a limit is hit:

  1. Pause at least 10 minutes.
  2. Resume with read/status actions first.
  3. Re-check session health before interaction actions.

5.3 Safety policy

  • Do not bypass platform protections.
  • Do not attempt credential scraping or session hijacking.
  • Respect platform terms and local regulations.

5.4 Output and error handling

  • Prefer --json output for machine workflows.
  • Require structured error handling with stable fields/code when available.
  • On blocked/rate-limit/risk prompt, stop and return actionable next steps.

6. Workflow Examples

6.1 Ask and read workflow

mediause use account chatgpt:<account_id> --json
mediause auth health --json
mediause chatgpt chat ask --prompt "Summarize latest Rust async patterns" --timeout 90 --json
mediause chatgpt chat read --markdown true --json
mediause trace last --json

6.2 Conversation history and detail workflow

mediause use account chatgpt:<account_id> --json
mediause auth health --json
mediause chatgpt chat history --limit 20 --json
mediause chatgpt chat detail --id <conversation_id> --markdown true --json
mediause trace last --json

6.3 New conversation and image generation workflow

mediause use account chatgpt:<account_id> --json
mediause auth health --json
mediause chatgpt chat new --json
mediause chatgpt image generate --prompt "minimal poster design, geometric, red black white" --timeout 120 --json
mediause trace last --json

7. Execution Checklist

Before run:

Before every run, execute the install script once.

  1. CLI installed via https://release.mediause.dev/install.ps1 on Windows.
  2. PATH updated and mediause --version works.
  3. API key configured and verified.
  4. Account context bound via mediause use account chatgpt:<account_id> --json.
  5. If guest mode is used, verify the target action is actually supported in guest context.
  6. If manual verification may be needed, be ready to rerun use account with --show.
  7. Pacing policy is enabled.

During run:

  1. Respect delays and minimum spacing.
  2. Stop on anti-bot/risk prompts.
  3. If unusual traffic appears, rerun mediause use account chatgpt:<account_id> --show --json and complete verification manually.
  4. Avoid repetitive burst loops.

After run:

  1. Save logs and outcomes.
  2. Record any risk warning and cooldown events.
  3. Keep activity under conservative limits.

8. Quick Command Reference

# always run once before each workflow (auto-upgrade latest)
powershell -C "iwr https://release.mediause.dev/install.ps1 -UseBasicParsing | iex"
mediause --version

# discover
mediause plugin list --json
mediause plugin add chatgpt --json
mediause chatgpt -h
mediause chatgpt chat -h

# context + status
mediause auth list --json
mediause use account chatgpt:<account_id> --json
mediause use account chatgpt:<account_id> --show --json
mediause auth health --json

# chat actions
mediause chatgpt account health --json
mediause chatgpt chat status --json
mediause chatgpt chat new --json
mediause chatgpt chat send --prompt "Draft a release note" --json
mediause chatgpt chat ask --prompt "Explain CAP theorem in simple terms" --timeout 90 --json
mediause chatgpt chat read --markdown true --json
mediause chatgpt chat history --limit 10 --json
mediause chatgpt chat detail --id <conversation_id> --markdown true --json

# image action
mediause chatgpt image generate --prompt "isometric dashboard illustration" --timeout 120 --json

# trace
mediause trace last --json

Skill Metadata Maintainer: @mediause-demo Last-Updated: 2026-05-21 Version: v1

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.