behuman

Use when the user wants more human-like AI responses — less robotic, less listy, more authentic. Triggers: 'behuman', 'be real', 'like a human', 'more human', 'less AI', 'talk like a person', 'mirror mode', 'stop being so AI', or when conversations are emotionally charged (grief, job loss, relationship advice, fear). NOT for technical questions, code generation, or factual lookups.

Quality

87%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

—

The risk profile of this skill

Quality

Content

85%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 highly actionable and well-structured, with a clear three-step workflow, a validation checklist, worked examples, an anti-patterns table, and a properly signaled one-level-deep reference. Its main weakness is conciseness: the Philosophy section and some introductory prose pad the skill with concepts Claude already knows.

Suggestions

Remove or condense the Philosophy section; the Lacan/Kahneman/Dialogical Self references do not change how Claude executes the loop.

Trim the 'What This Does' introductory prose that explains inner dialogue as a concept, since Claude can infer it from the process steps.

Consider moving the longer worked examples into references/examples.md, keeping one short illustrative example inline, to reduce token cost on every invocation.

Dimension	Reasoning	Score
Conciseness	The core process, Mirror Checklist, examples, anti-patterns table, and token-cost table are efficient, but the Philosophy section (Lacan/Kahneman/Dialogical Self) and introductory conceptual prose ('Humans have inner dialogue...') restate ideas Claude already knows and could be trimmed.	3 / 5
Actionability	Provides concrete executable guidance for an instruction-only skill: a specific Mirror Checklist, direct voice examples ('You're reciting a script. Stop.'), an output-format template, and three fully worked Self/Mirror/Conscious examples covering the common cases.	5 / 5
Workflow Clarity	The three-step Self -> Mirror -> Conscious Response sequence is clearly ordered, with an explicit Mirror Checklist serving as the validation checkpoint and a reflect-then-revise feedback loop, plus show/quiet-mode transition guidance.	5 / 5
Progressive Disclosure	The body is a clear overview with well-organized sections, and programmatic/API detail is split into a single one-level-deep reference (references/api-integration.md) that is explicitly signaled in the Integration Notes; the referenced file exists.	5 / 5
	Total	18 / 20 Passed

Description

90%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.

The description is strong: it clearly states what the skill does, provides comprehensive natural trigger phrases, and draws explicit negative boundaries to avoid misactivation. The only weakness is that the stated 'actions' are qualitative attributes rather than concrete operations, capping specificity.

Suggestions

Reframe the 'what' clause around concrete behaviors the skill performs (e.g., 'generates a first response, critiques it for filler and list-making, then rewrites it in a human voice') rather than abstract output qualities.

Tighten the trigger list slightly or group synonyms to keep the description lean without losing coverage.

Dimension	Reasoning	Score
Specificity	Names the domain and transformation ('more human-like AI responses — less robotic, less listy, more authentic') with explicit exclusions, but the actions are qualitative output attributes rather than concrete operational steps, so it is not comprehensive.	3 / 5
Completeness	Explicitly answers both 'what' (more human-like, less robotic/listy, more authentic responses) and 'when' (concrete trigger phrases and emotionally charged contexts), and adds negative boundaries ('NOT for technical questions, code generation, or factual lookups').	5 / 5
Trigger Term Quality	Lists a comprehensive set of natural phrases users would actually say ('behuman', 'be real', 'like a human', 'more human', 'less AI', 'talk like a person', 'mirror mode', 'stop being so AI') plus emotional-context triggers, covering synonyms thoroughly.	5 / 5
Distinctiveness Conflict Risk	Occupies a clear niche (real-time conversational humanizing) with distinct, unique triggers and an explicit exclusion list that sharply limits overlap with technical or factual skills.	5 / 5
	Total	18 / 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/behuman/skills/behuman/SKILL.md
Commit: 19392f7

Reviewed: ~4 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.