Description
santa-method is an agent skill from the ECC collection by affaan-m, listed on skills.sh. It is a multi-agent adversarial verification method. Two independent reviewers, who share no context with each other or with the author, must both pass the same rubric before output is shipped, and when one fails, the output is fixed and reviewed again until both pass. The skill's reasoning is that a single agent reviewing its own work shares the biases and gaps that produced the work, so independent reviewers are needed to catch systematic errors. It is for output that will be published, deployed or used by end users, for compliance, regulatory or brand constraints, for code that ships without human review, for content where accuracy matters, for batch generation where spot checks miss patterns, and for text with elevated hallucination risk, such as claims, statistics, API references or legal language. The skill says not to use it for internal drafts, exploratory research or tasks with deterministic checks, which build, test and lint pipelines handle better. A skill is a SKILL.md file of instructions that an AI agent loads when a task calls for it. It is free and open source, and the skills.sh listing shows all three security audits passing.…
Highlights
- Two independent reviewers with the same rubric
- Fix and re-review until both pass
- Targets hallucinated claims and statistics
- Not for drafts or deterministic checks
Ratings & Reviews
No ratings yet
Be the first to rate santa-method.
People also visit
santa-method FAQ
What does the santa-method skill do?
It runs output through two independent reviewers who apply the same rubric, and only ships it when both pass, fixing and re-reviewing as needed.
Why two reviewers?
The skill says one agent reviewing its own output shares the same biases and gaps, while independent reviewers do not.
When should I not use it?
For internal drafts, exploratory research, or tasks with deterministic checks that build, test or lint pipelines can cover.
How do I install it?
Run: npx skills add https://github.com/affaan-m/ecc --skill santa-method
Is it free?
Yes. It is free and open source.
Are there security findings?
The skills.sh listing shows Gen Agent Trust Hub, Socket and Snyk all passing.
What are the best santa-method alternatives?
Popular alternatives to santa-method include blueprint, claude-devfleet and council. See all santa-method alternatives