sk Skill · affaan-m
santa-method
Multi-agent adversarial verification with convergence loop. Two independent review agents must both pass before output ships. Use when output must clear two independent adversarial reviewers before it ships.
Open on skills.sh ↗read 2026-09-15
- installs 8w
- 1,709
- 30-day movement
- starts with the next reading
- Related entries
- 6
- Connections
- 0
pythonbashJavaScript
Santa Method is an agent skill that adds a post-generation verification layer: two independent adversarial review agents check output against a rubric, and both must pass before it ships. It includes a convergence loop where failing output is fixed and re-reviewed by fresh agents to avoid anchoring bias.
Reach for it when output must clear two independent adversarial reviewers before it ships.
Use it to
- Verify publishable or user-facing output before release
- Enforce compliance, regulatory, or brand constraints
- Design objective pass/fail rubrics for dual review
- Batch-verify large outputs with stratified sampling
- Run sequential inline review when subagents are unavailable
For Agent users shipping high-stakes generated output
- Host repository
- affaan-m/ECC
- Installs, lifetime
- 2,900
- Installs, 8 weeks
- 1,709
- Host stars
- 259k
- Host language
- JavaScript
topicsverificationmulti-agentreviewquality-assurancerubricconvergence-loop