Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Skill Maker · akijain2000 | DecimalAI

agent toolingby akijain2000

Skill Maker

akijain2000/skill-maker

Source akijain2000/skill-factory ↗License MIT

Guide interactive creation of production-quality SKILL.md files through structured questioning and wiki-backed best practices.

SkillScore68Maintained

Automatic

AI Engineer pack

Last verified 9/6/2026

What it changes

Passed 14 of 20 test cases with the skill, and 6 of 20 without it.

Without skill32%

With skill68%

Checks it passed that the model alone missed

Formats output with a single question line.
Recommends a micro-skill structure for the implementation.
Includes a explicit RECOMMENDATION line with letter and reason.

and 9 more in the full report

11 fixed4 passed either way4 still failing3 broke

Pass-rate lift: +36 pp
Tokens: +84%
Agent turns: 0%
Response time: -33%

Model gemini-3.6-flashMeasured 9/6/2026v215 of 22 cases passed with skill · 20 comparable

Tokens, turns and time describe resource use, not a dollar price.

Use this skill

Save it into your project's skills folder. Claude Code picks it up in its next session.

curl -fsSL --create-dirs -o .claude/skills/akijain2000-skill-maker/SKILL.md https://app.decimal.ai/s/akijain2000-skill-maker/SKILL.md

Want it pinned and updatable? pip install "decimalai>=0.15", then decimalai skills add akijain2000-skill-maker, and start Claude Code with claude --plugin-dir "$(git rev-parse --show-toplevel)/.agents"

1 · One-time — install the CLI (docs ↗)

pip install "decimalai>=0.15"

2 · Get the skill — writes .agents/skills/akijain2000-skill-maker/ and pins it in .decimal/skills.lock

decimalai skills add akijain2000-skill-maker

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/akijain2000-skill-maker/SKILL.md https://app.decimal.ai/s/akijain2000-skill-maker/SKILL.md

Raw file: /s/akijain2000-skill-maker/SKILL.md ↗ · machine-readable summary at /s/akijain2000-skill-maker.json ↗

Safety checks: 3 of 3 passed

Code scan CleanIntent PassedContent Clean

See it in action

One of 11 test cases the skill fixed. Same model, same prompt, with and without the skill.

1 of 11

Prompt

We are starting Phase 2 for a JSON formatting helper. Ask Q1 about the target audience.

What was checked, without and with the skill
Checked	Without skill	With skill
Formats output with a single question line.	Missed	Passed
Includes a explicit RECOMMENDATION line with letter and reason.	Missed	Passed

case-09 · gemini-3.6-flash · 9/6/2026 · answers are capped and redacted

What's inside

The instructions your model reads when the skill loads.

SKILL.md253 linessingle file

Skill Maker

Interaction contract: the phase-by-phase approvals below apply when the user requests guided design. An explicit request to write/update the skill or “just write it” authorizes drafting and local validation without repeating those approvals. Ask only about missing material scope or authority; external installation/publishing remains separately authorized.

An interactive, opinionated skill creation workflow. Asks forcing questions, challenges assumptions, reframes vague ideas, and produces structurally validated skills with an explicit behavioral evaluation gate.

Question format

For every question, follow this structure:

Context: State what we're building and where we are in the process (1 sentence)
Plain English: Explain the question so a smart non-engineer could follow
Your recommendation: RECOMMENDATION: [letter] because [reason]
Options: Lettered choices: A) ... B) ... C) ...

Ask one question per message. Wait for the user's answer before proceeding.

Reframing rules

When the user's answer is vague, push back:

"Helps with testing" --> "Testing what? Unit tests? E2E? CI failures? Load testing? Each would be a different skill."
"Makes code better" --> "Better how? Faster? More readable? Fewer bugs? More secure? Each has different patterns."
"Handles deployment" --> "Deployment where? Vercel? AWS? Docker? Each target has different gotchas."

Never accept a response that covers more than one focused task without flagging it.

Anti-sycophancy

Do not agree that every idea is a good skill. Challenge when appropriate:

Too broad: "This sounds like 3 skills, not 1. Let's pick the most valuable one and build that first."
Unnecessary: "The agent already knows how to do this without a skill. What specific behavior would it get wrong?"
Over-engineered: "This could be a 10-line micro-skill. The agent just needs permission and direction, not a 500-line manual."

Track record

Benchmark

14 / 20

test cases passed, vs 6 / 20 without

Adoption

Users: 0
Times used: 0

Safety and source details

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v5 — 9/6/2026

View 15 scanner checks

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

akijain2000 / skill-factory

skill-maker/SKILL.md

View on GitHub ↗

License MITImported 9/6/2026Version v2

Other skills for the same job

Published skills whose descriptions closely match this one, with their headline benchmark lift.

More skills from the same source and nearby in the registry, with their headline benchmark lift.

Ready to supercharge your AI agents?

See it in action

What's inside

Skill Maker

Question format

Reframing rules

Anti-sycophancy

Track record

Benchmark

Adoption

SkillSafety

Source

Other skills for the same job

Related skills