Tessl • The package manager for agent skills

SkillPublic

zero-hallucination-coder

Runs a disciplined Discuss → Map → Decompose → Execute → Verify loop that grounds code in verified structure — no invented APIs, no assumed imports, no placeholder code — with a lazy-senior-dev YAGNI ladder that deletes unnecessary code before it is written. Use when a coding task is high-stakes, complex, or spans existing code (auth, databases, migrations, multi-file features), or when the user explicitly asks to plan carefully before coding, avoid hallucinated code, or work rigorously. Not for trivial edits, typos, or throwaway one-off scripts — those do not need the full loop.

Quality

90%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

81%Weight 40%Scale 1-5

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

The body is a highly actionable, clearly sequenced process skill with strong validation checkpoints and feedback loops throughout. Its main cost is token efficiency: attribution content and repeated scope guidance inflate the file without adding operational value.

Suggestions

Remove or compress the Credits & Inspiration section to a single line (or move it to a reference file) — the attribution table adds ~12 lines with no operational value.

Consolidate the overlapping scope guidance in 'When to invoke' and 'When to Short-Circuit' into one decision table; the same high-stakes vs. trivial distinction is currently stated three times.

Trim explanatory asides for concepts Claude already knows (e.g., the definition of 'context rot' and the stdlib enumeration in Ponytail rung 2) down to the rule itself.

Dimension	Reasoning	Score
Conciseness	The core loop is tight, but the ~230-line body carries padding: the Credits & Inspiration table contributes nothing operationally, scope guidance is restated across 'When to invoke', 'When to Short-Circuit', and the description, and parentheticals explain concepts Claude already knows (e.g., 'context rot — the silent quality degradation as the context window fills').	3 / 5
Actionability	Every phase ships a copy-paste-ready template (CODEBASE MAP, IMPLEMENTATION PLAN, PONYTAIL CHECK, SELF-REVIEW, HANDOFF, VERIFICATION REPORT) plus concrete right-sizing examples ('Add validateToken(token: string): boolean to AuthService'); as an instruction-only skill, the guidance is fully executable with no vague steps.	5 / 5
Workflow Clarity	The five phases are strictly sequenced with explicit validation gates at every boundary: user-confirmed Situation Summary, no unresolved UNKNOWN flags before coding, per-story self-review verdicts, user confirmation before the next story, and a final verification report with a feedback loop (gap → micro-story → re-run Phase 4).	5 / 5
Progressive Disclosure	A single well-organized file with clear headers, tables, and no nested references, so navigation is easy; but at ~230 lines, secondary material (the credits table, the detailed Ponytail ladder) stays inline rather than being split out, leaving minor organization gaps.	4 / 5
	Total	17 / 20 Passed

Description

96%Weight 40%Scale 1-5

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

A strong description: concrete, third-person, with explicit what/when guidance and natural trigger phrases including negative scope. The only weakness is that its trigger territory ('high-stakes, complex coding') is broad and could overlap with related coding-discipline skills.

Dimension	Reasoning	Score
Specificity	The description lists multiple concrete actions — 'Runs a disciplined Discuss → Map → Decompose → Execute → Verify loop', 'no invented APIs, no assumed imports, no placeholder code', and a 'YAGNI ladder that deletes unnecessary code before it is written' — comprehensively covering the skill's behavior in third-person voice.	5 / 5
Completeness	It explicitly answers both what the skill does (the five-phase loop, grounding rules, YAGNI ladder) and when to use it, with an explicit 'Use when...' clause containing concrete trigger phrases and clear exclusions.	5 / 5
Trigger Term Quality	It includes natural phrases users would actually say — 'plan carefully before coding', 'avoid hallucinated code', 'work rigorously', 'high-stakes' — plus concrete domain terms (auth, databases, migrations, multi-file features) and negative scope ('Not for trivial edits, typos, or throwaway one-off scripts').	5 / 5
Distinctiveness Conflict Risk	The zero-hallucination/YAGNI framing and negative scope make it mostly distinct, but 'a coding task is high-stakes, complex, or spans existing code' is broad enough to overlap with general coding, planning, and code-review skills.	4 / 5
	Total	19 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: alirezarezvani/claude-skills
Path: engineering/zero-hallucination-coder/skills/zero-hallucination-coder/SKILL.md
Revision: 27cfcd0

Reviewed: 2 hours ago

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.

zero-hallucination-coder

Table of Contents

Is this your skill?