Skip to content
Agentmart
Submit a tool

slate-critics

Use after Export as SLATE desk 12 Critics.

Skill by Oncorporation

IndexedMIT
Write a review

Chalk Labs

Need help deploying slate-critics?

Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.

Book a call

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/oncorporation-slate-critics

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Use after Export as SLATE desk 12 Critics. Review the previous cut for dialogue, direction, continuity, scene, picture QA, or audio QA. First use presents a per-subagent tool menu. Triggers on critics desk, review the cut, Picture QA, Audio QA, hold the cut.

Use cases: testing, agent-building

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/slate-critics

What's inside

Skill name
slate-critics
Description
Use after Export as SLATE desk 12 Critics. Review the previous cut for dialogue, direction, continuity, scene, picture QA, or audio QA. First use presents a per-subagent tool menu. Triggers on critics desk, review the cut, Picture QA, Audio QA, hold the cut.
License
MIT
Instructions
93 lines in SKILL.md · view on GitHub

Agent Skills spec issues

  • files: referenced but missing: templates/critics/tools.md

Files (3)

  • SKILL.md
  • menus.md · references/
  • tools.md · references/

SKILL.md (first lines)

# Critics desk (12)

Review desk. Not a Producer. Notes only.

Producers are desks 2–11. Critics never edit Fountain, `beats.md`, `shots.md`, boards, or `assets.md`.

Showrunner routes fail notes to the owning Producer. Human holds the previous cut. No land and no quality-band bump without the writer.

## When to run

After Export writes a cut. Skip Picture QA unless export type is `video`. Skip Audio QA unless type is `audio` or the video cut has a track.

## Cut record

Write under the output root:

```
critics/tools.md
critics/{cut-id}/manifest.md
critics/{cut-id}/dialogue.md
critics/{cut-id}/direction.md
critics/{cut-id}/continuity.md
critics/{cut-id}/scene.md
critics/{cut-id}/picture.md    # video only
critics/{cut-id}/audio.md      # audio, or video with a track
```

`{cut-id}` = `{export-type}-{band}-{date}` or the Export folder name. Point `bible.md` **Previous cut** at this id.

Manifest fields: export type, band, files reviewed, scouts run, tools used, open holds.

## First use — tool menu

Each scout is editable (`agents/critic-*.md` plus `critics/tools.md`).

On first use of a scout (that scout has no `tools:` line, or `tools: unset`):

1. Load `references/menus.md`.
2. Show **only that scout's** menu. Fit the job. Do not list Picture tools on Dialogue.
3. Wait. Do not review yet.
4. Write picks into `critics/tools.md`.
5. Next turn, run those tools.

If the writer edits `critics/tools.md` or a `agents/critic-*.md` charter, obey the file. Re-show a menu only when they say `reset critic tools` or a scout is still `unset`.

`llm-craft` is always on. Other tools are opt-in.

Copy `templates/critics/tools.md` to `critics/tools.md` if missing.

## Run order

1. Confirm previous cut path + export type + band.
2. Gate tools (menu if unset).
3. Spawn the matching scouts. Parallel notes allowed. One report file each.
4. Fold pass/fail into `manifest.md`.
5. Hold for the writer. Present the cut + scores + fails.
6. On writer accept — lock story at this band. On patch — list owning Producers. On bump — only if story scouts passed and the writer said so.

## Verdict

Each scout report: `pass` | `fail` | `hold`. Evidence first (tool output), then craft note. Cite Fountain lines, `sh_` ids, or timestamps.

Story scouts (Dialogue, Direction, Continuity, Scene) must pass before any video band above 10 or any audio tier bump.

## Quality bands (video)

- 0–25 simple animation / storyboard
- 26–50 illustrated
- 51–75 CGI
- 76–100 cinematic photoreal

Default ladder: band 10 → story lock + human accept → 65 or 90.

Text has no picture band. Audio uses production tier via Audio QA, not photoreal scores.

## Routing

| Scout fail | Route to |
|---|---|
| Dialogue | Dialogue Producer |
| Direction | Camera and/or Scene |
| Continuity | Casting, Architect, or Scene |
| Scene | Drama and/or Scene |
| Picture QA | Export only |
| Audio QA | Export only |

## Load

- `references/tools.md` — repos and what each tool measures
- `references/menus.md` — first-use menus
- matching `agents/critic-*.md`

From Oncorporation/slate-agent, MIT.

More from Oncorporation/slate-agent

Signals

Stars
0
Forks
0
Open issues
0
Last commit (this folder)
Sep 13, 2026 (1 month ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from Oncorporation

Done for you

Rather not set it up yourself?

Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.

See how we build AI agents →Read the case studies

Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.

launches shipped
50+
launches shipped
media placements
500+
media placements
founders on every account
2
founders on every account

Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block

Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.

Get slate-critics running in your business

Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.

Goes to the Chalk Labs founders, not the newsletter. Privacy

Need help deploying slate-critics?

CallGet help
Cite this page

slate-critics skill. Agent Mart. https://agentmart.dev/skills/oncorporation-slate-critics (last updated 2026-10-24)