Agui Author

What it does

When your agent uses it

• You want the operator to see a decision
• A status readout instead of scrolling terminal text

Example prompts

• “/agui-author”

Requirements

• Python 3
• A Bash shell

Workflow steps

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

1. 1Emitting to a disabled surface — if CAO_AGUI_ENABLED is unset, emit_ui
2. 2Props over 8 KB are rejected — the tool raises a ValueError and nothing
3. 3No HTML sink exists — strings in props render as plain text. Attempting to
4. 4One intent per meaningful moment — emitting a progress card on every
5. 5approval_card is display-only today — it gives the operator an
6. 6Off-list components are refused server-side — the allow-list is fixed

What it can do on your machine

Read from SKILL.md and the folder at commit 089c53c. 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 1 file in scripts/ (Shell), which the agent can run.

  Shell commands in SKILL.md call:

  • curl
  • uv

  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 uv, 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

Agui Author loads about 2k tokens when it runs, and up to ~3.9k if it reads all its reference files. Until then it costs about 103 tokens; SKILL.md has 809 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~103

When it runs · the whole SKILL.md, loaded when a task matches

~2k

With references · SKILL.md plus every file in references/, read only if the agent opens them

~3.9k

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

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); the scripts in this folder are not scanned.

SKILL.md

The full file from awslabs/cli-agent-orchestrator at commit 089c53c, republished under its Apache-2.0 licence (© awslabs). 809 words, ~1,963 tokens.

name

agui-author

description

Author live dashboard UI from an agent via the `emit_ui` MCP tool. Emit one of six allow-listed components (approval_card, choice_prompt, diff_summary, progress, metric, agent_card) with JSON props and it renders in any AG-UI client watching the fleet. Use when you want the operator to see a decision, a diff, or a status readout instead of scrolling terminal text. Arbitrary HTML/markup is refused.

emit_ui(component: str, props: dict) -> {"ok", "event_id", "component"}

# Gate a risky action on human approval.
emit_ui("approval_card", {
    "title": "Deploy to production?",
    "detail": "3 files changed, 1 DB migration",
    "risk": "high",
})

# Ask the operator to choose.
emit_ui("choice_prompt", {
    "question": "Which base branch?",
    "choices": [{"label": "main", "value": "main"},
                {"label": "release", "value": "release"}],
})

# Summarize a change set.
emit_ui("diff_summary", {
    "title": "Refactor auth",
    "files": [{"path": "security/auth.py", "additions": 74, "deletions": 3}],
})

# Show progress / a metric / your identity.
emit_ui("progress", {"label": "Indexing repository", "value": 0.42})
emit_ui("metric", {"label": "tokens used", "value": 12840, "unit": "tok"})
emit_ui("agent_card", {"name": "reviewer", "provider": "claude_code", "status": "working"})

# 1. Server with the surface on
CAO_AGUI_ENABLED=true uv run cao-server

# 2. Watch the stream (SSE frames print as they arrive)
curl -N 'http://localhost:9889/agui/v1/stream'

# 3. Emit from anywhere (the MCP tool does exactly this)
curl -sX POST http://localhost:9889/agui/v1/emit_ui \
  -H 'Content-Type: application/json' \
  -d '{"component":"progress","props":{"label":"demo","value":0.5}}'

© awslabs, Apache-2.0. 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 4 other files (scripts, references) in skills/agui-author of awslabs/cli-agent-orchestrator.

• SKILL.md
• evals/evals.json
• references/l2-constructs.md
• references/run-plane.md
• scripts/validate.sh

Open the folder on GitHubat commit 089c53c