santa-method

Multi-agent adversarial verification with convergence loop. Two independent review agents must both pass before output ships.

Quality

60%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Risky

Do not use without reviewing

Fix and improve this skill with Tessl

tessl review fix ./skills/santa-method/SKILL.md

Quality

Content

62%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body delivers a clear, well-sequenced multi-agent verification workflow with strong feedback loops and a concrete reviewer prompt. It is weakened by length, undefined helper functions in the orchestration code, and an absence of progressive disclosure for a skill this size.

Suggestions

Define or stub the helper functions (dedupe, fix_agent.execute, generate, ship, escalate_to_human) so the orchestration code is executable rather than illustrative.

Move domain-specific rubric extensions, Metrics, and Cost Analysis into separate reference files linked one level deep from SKILL.md to apply progressive disclosure.

Tighten or trim the Cost Analysis and Integration-with-Other-Skills sections, which add tokens without materially guiding execution.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient and avoids explaining basic concepts Claude already knows, but at ~300 lines it includes tightenable sections (Cost Analysis, Integration with Other Skills, the core-insight rationale) that pad the token budget.	2 / 3
Actionability	Provides a concrete reviewer prompt template and clear verdict/loop logic, but the orchestration code relies on undefined helpers (dedupe, fix_agent.execute, generate, ship, escalate_to_human, classify_failure, batch_fix), so it is not copy-paste executable.	2 / 3
Workflow Clarity	Four phases are clearly sequenced with an explicit verdict-gate validation checkpoint, a fix→revalidate feedback loop, fresh-reviewer-per-round discipline, and an iteration cap with escalation — matching the score-3 anchor.	3 / 3
Progressive Disclosure	Sections are well-organized, but the skill is a monolithic ~300-line SKILL.md with no bundle files; content like domain-specific rubric extensions, metrics, and cost analysis is inline that could be split into one-level-deep references.	2 / 3
	Total	9 / 12 Passed

Description

57%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description conveys a distinctive, well-scoped mechanism but stops short of explicit activation triggers and relies on somewhat technical jargon over natural user phrasings. It answers 'what' well but not 'when'.

Suggestions

Add an explicit 'Use when…' clause listing concrete trigger situations (e.g., before publishing/deploying output, when hallucination risk is high, for compliance-sensitive content).

Replace or supplement jargon like 'adversarial verification' and 'convergence loop' with natural terms users would actually say ('independent review', 'check before shipping', 'fix until it passes').

Name a couple more concrete actions (e.g., 'spawn two reviewers, gate output on both passing, loop fixes until convergence') to lift specificity.

Dimension	Reasoning	Score
Specificity	Names the domain ('multi-agent adversarial verification') and a concrete mechanism ('Two independent review agents must both pass before output ships'), but does not enumerate multiple specific actions like the score-3 anchor.	2 / 3
Completeness	Clearly states what the skill does, but lacks a 'Use when…' clause or equivalent explicit trigger guidance for when Claude should invoke it, so completeness is capped at 2 per the rubric guidelines.	2 / 3
Trigger Term Quality	Contains relevant terms ('verification', 'review agents', 'output ships') but leans on jargon ('adversarial verification', 'convergence loop') and omits common natural phrasings users would say like 'check', 'verify', or 'quality review before shipping'.	2 / 3
Distinctiveness Conflict Risk	The 'multi-agent adversarial verification with convergence loop' and 'two independent review agents must both pass' signature carves a clear niche that is unlikely to trigger for unrelated review skills.	3 / 3
	Total	9 / 12 Passed

Validation

93%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
metadata_version	'metadata.version' is missing	Warning

	Total	15 / 16 Passed

Repository: affaan-m/ECC
Commit: 754b8dd

Reviewed: 3 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.