Name: human-gate
Rating: 78.4 (1 reviews)
Author: alirezarezvani

human-gate

Runs the human-verification lane of an agent loop, and proves review happened before work is called done. Builds a single-file HTML review page, collects batched feedback as a structured artifact instead of chat prose, and runs a gate that refuses to close while a BLOCKER is open, the reviewer is unnamed, or nobody has reviewed at all. Use when a plan, spec, RFC, report, landing page, migration, or any irreversible action needs human sign-off before shipping, or on requests such as 'get sign-off', 'have someone check this', 'hold until reviewed', 'needs approval first'. NOT for making AI text sound human (use content-humanizer or behuman). NOT for reviewing code diffs (use md-review or code-reviewer).

Quality

98%

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

The canonical home for this skill is human-gate in alirezarezvani/claude-skills

Quality

Content

96%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.

A tight, highly actionable body: executable commands, explicit gate/validation logic, and disciplined outward references. The only soft spot is progressive disclosure, where the signaled reference files are not present in the reviewed bundle to verify.

Suggestions

Ship the referenced bundle (references/, assets/, audit/human-review-2026-08/) alongside SKILL.md so the one-level-deep pointers can be verified and progressive_disclosure can score 5.

Consider moving the G1–G7 gate-rules table and the EDIT/severity definitions into references/ to slim the body further, keeping SKILL.md a pure overview.

Dimension	Reasoning	Score
Conciseness	Lean and dense throughout — no padding explaining what HTML, code review, or feedback batching is; every line earns its place and assumes Claude's competence. The one time-sensitive item (human-review@0.6.0, 2026-08) is tied to a security rationale, not loose deprecation prose.	5 / 5
Actionability	Provides copy-paste-ready shell commands with exact paths and flags, explicit exit-code semantics ('exit 2 = NOT done'), a '--sample' run of the whole loop, and a worked sidecar example — fully executable covering the common cases.	5 / 5
Workflow Clarity	The open → status → collect → close loop is explicitly sequenced with validation checkpoints: a gate that refuses to close (G1–G7), round-cap exhaustion that escalates, and an integrity guard (G7 typo-downgrade). This skill is itself a validation workflow, and its feedback loops and checklists are concrete.	5 / 5
Progressive Disclosure	Clear sectioned overview with well-signaled one-level-deep pointers (references/, assets/, audit/human-review-2026-08/) and a guarded 'if it exists' ctx file, but the referenced bundle files are absent from this review directory so they cannot be confirmed, and the body is fairly content-dense.	4 / 5
	Total	19 / 20 Passed

Description

100%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.

An exemplary description: concrete capabilities, natural trigger phrases, explicit what-and-when guidance, and explicit anti-triggers naming sibling skills. No verbosity or over-claims.

Dimension	Reasoning	Score
Specificity	Names multiple concrete actions — 'Builds a single-file HTML review page,' 'collects batched feedback as a structured artifact,' 'runs a gate that refuses to close while a BLOCKER is open' — with comprehensive coverage and no vague filler.	5 / 5
Completeness	Explicitly answers both 'what' (builds review page, collects batched feedback, runs a gate) and 'when' via a concrete 'Use when...' clause with multiple trigger scenarios; not below 5 since neither half is missing or weak.	5 / 5
Trigger Term Quality	Includes natural phrases users actually say ('get sign-off', 'have someone check this', 'hold until reviewed', 'needs approval first') alongside domain nouns (plan, spec, RFC, report, landing page, migration), giving comprehensive coverage.	5 / 5
Distinctiveness Conflict Risk	Carves a clear niche and adds explicit NOT-for boundaries naming competing skills (content-humanizer, behuman, md-review, code-reviewer), minimizing trigger overlap.	5 / 5
	Total	20 / 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: .gemini/skills/human-gate/SKILL.md
Commit: 19392f7

Reviewed: ~7 hours 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.