Four-Color Cards (Fact · Explanation · Risk · Action)
Turns a parsed document plus a user intent into four auditable card types — blue facts, green explanations, yellow risks, and red actions — with mandatory source citation.
Data & AnalyticsAdvanced★ 6,083⑂ 585AI score 6/10Last updated: Sep 2, 2026
What it does
- Takes structured markdown and an analysis intent, then produces four card types:
- Blue (fact): traceable statements with source snippets
- Green (explanation): mechanism and background behind each fact
- Yellow (risk): over-claims, uncertainty, potential bias
- Red (action): concrete next-step recommendations
- Enforces a citation chain: green cites blue, red cites green/blue, and uncited claims are rejected.
- Defines graceful degradation: without an LLM it falls back to a rule engine and labels cards accordingly; review timeouts mark cards
pending_review. - Ships a runnable entrypoint with three stages (
extract,review,propose) writing JSON artifacts.
Who it's for
- Analysts producing research, financial, or legal summaries that must stay verifiable.
- Engineers building multi-agent extract → critique → recommend pipelines.
- Teams that need to audit AI output for overstated conclusions.
Examples
- Feed a literature survey and run
--stage extractto get blue fact cards with quoted evidence. - Run
--stage reviewto generate green cards exposing reasoning gaps and over-claims. - Run
--stage proposeto get red action cards, saved toexamples/snse_survey/skill_outputs/four_color_propose.json.
· · · Install guide · · ·
Try it now, no install
Paste this into Claude to use the skill without installing anything.
Read the instructions in this file and follow them to help me: https://raw.githubusercontent.com/anbeime/skill/HEAD/antinet-agentteams/skills/four-color-cards/SKILL.md What I want: (describe your task here)
If Claude can't open the link, open it yourself and paste the contents instead.
↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.
Install in the Claude app (no terminal)
- Download the ZIP with the button below.
- In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
- Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
Install in Claude Code
Let Claude do it — paste this into Claude Code
Install the skill I found on Claude Skill Mart. Copy the antinet-agentteams/skills/four-color-cards folder from the GitHub repo anbeime/skill into my ~/.claude/skills/four-color-cards/. When it's done, tell me in one line what this skill can do.
Install with a command instead
git clone https://github.com/anbeime/skill.git && mkdir -p ~/.claude/skills && cp -r skill/antinet-agentteams/skills/four-color-cards ~/.claude/skills/⚠ This is a third-party skill. Check the source repository before installing.
- Open a terminal and go to your working directory.
- Clone the repository:
git clone https://github.com/anbeime/skill.git - Create the skills folder:
mkdir -p ~/.claude/skills - Copy the skill:
cp -r skill/antinet-agentteams/skills/four-color-cards ~/.claude/skills/ - If you want the Python pipeline, install the repo's dependencies and configure your LLM API key.
- Restart Claude Code and try: "Build four-color cards for this document."
- Script example:
python skills/four-color-cards/scripts/run_four_color_cards.py --stage extract