Santa Method
Multi-agent adversarial verification: two independent reviewers with the same rubric must both pass before output ships, with fix-and-re-review convergence and a human escalation cap.
- What
- Multi-agent adversarial verification: two independent reviewers with the same rubric must both pass before output ships, with fix-and-re-review convergence and a human escalation cap.
- Cost
- Free
- Needs
- Use "Santa Method" with your Muse.
- Install
- Copy the installer prompt below into your Muse — your agent does the rest.
Curated by Skill Harbor: an adversarial verification framework built on one insight, make a list and check it twice. A single agent reviewing its own output shares the same biases, knowledge gaps, and systematic errors that produced the output; two independent reviewers with no shared context break that failure mode. The architecture runs three phases. Phase 1, make a list: the generator agent produces the deliverable. Phase 2, check it twice: two independent reviewers score the output against the same rubric, blind to each other. Phase 3, naughty or nice: the verdict gate ships the output only if both reviewers pass; otherwise the output goes back for fixes and the loop converges through re-review, with a human escalation cap so the loop cannot spin forever on an unfixable disagreement. Designed for gating publishing, production deploys, compliance or brand-sensitive content, and hallucination-prone claims before they ship. Explicitly not for internal drafts, exploratory research, or tasks with deterministic verification, where build and test pipelines are the right tool. By Ronald Skelton of RapportScore.ai, via the affaan-m/ECC repository (MIT), listed here with credit to its creator. Honest caveats: dual review costs roughly twice the tokens; the rubric is everything, a weak rubric just double-certifies weak output; the escalation cap must be real, or the loop never ends. Skill Harbor never reviews the code, review it yourself before use.
Version:
Install
Copy the install package below, then paste it into MuseThe install prompt below already includes the vetting steps: your agent follows the community checklist before installing anything with executable code. Want more?
Use "Santa Method" with your Muse. Prerequisites: none to install. Pure guidance; costs roughly twice the tokens of a single review, so reserve it for what ships. 1. Open the skill: https://github.com/affaan-m/ECC/blob/main/skills/santa-method/SKILL.md and copy the full SKILL.md text. 2. Paste it into a chat with Muse and add: "Run the Santa Method on [deliverable] with this rubric: [paste or describe the rubric]." 3. Let the generator produce, the two reviewers score independently, and the gate decide. If it fails, fix and re-review; escalate to a human at the cap. Tip: write the rubric first and make it specific; a vague rubric is the most common way to waste the double review. Safety: a skill is plain-text instructions; it runs nothing by itself. The human escalation cap is yours to enforce.
Saved to your recent installs. Find it anytime on /connect.
Questions
How do I install a build?
Every product page includes a copy-paste install prompt. Paste it into your Muse and it sets the build up for you — no manual configuration.
Where does my money go?
Straight to the seller. Skill Harbor never processes payments: checkout happens on the seller’s own page, usually Stripe.
What does the ✓ next to a creator’s name mean?
It means we confirmed the identity of the person behind the listing. It says nothing about the code itself — always check a build before installing it.