Skip to content
Get packs

Cross-Model Memo Review

Has several AI models independently critique a high-stakes board memo and reconcile their votes

Source Download skill
Repository
alirezarezvani/claude-skills
Stars
29,149
Licence
MIT
Files
1 file

What it does

When a founder or executive has a high-stakes memo, such as a fundraise term sheet or a layoff plan, this skill tells the agent to send the memo to several different AI models (Claude, plus OpenAI's Codex and Google's Gemini if they are set up). Each model is asked to list its top concerns and supports and vote APPROVE, REJECT or DEFER. The agent then groups points that multiple models agree on, surfaces the disagreements as questions for the founder, and saves a markdown report. If only Claude is available, it runs three differently prompted Claude passes instead and labels the result as weaker.

Great for

  • Founders or executives about to make an irreversible decision, such as a major pivot, acquisition or fundraise
  • Teams that want an independent sanity check on a board memo before the real board meeting
  • People who already have API keys or CLIs for several AI providers and want a multi-model comparison

Skip it if

  • Routine or reversible decisions, where the cost of multiple reviews outweighs the benefit
  • Legal, tax or regulatory sign-off: the skill itself says it is not a replacement for outside counsel or a real board, and model judgements can be wrong or overconfident
  • Users who expect the models to be independent experts on their specific market; the skill assumes model biases differ but does not verify that

Try asking

  • “Run /c-level-agents:cross-eval on my draft memo recommending we accept this term sheet”
  • “Cross-review our layoff proposal memo before we announce it to the team”
  • “Get three independent reviews of our plan to enter the German market and list the open questions”
Step-by-step guide: how to use Cross-Model Memo ReviewInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 1 original files, the MIT licence and an attribution note.

    cross-eval.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip cross-eval.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip cross-eval.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Provide the full memo text or a link so every model reviews the same material; the output template records the memo link.
  • Read the divergent concerns first, since disagreement between models is the main signal the skill produces.
  • Treat a Claude-only run as suggestive rather than conclusive, as the skill itself advises.

What's inside

  • SKILL.md

Redistributed unmodified from alirezarezvani/claude-skills under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.