Ready to supercharge your AI agents?

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

Get Started Free →

Skill Evolver · stijnman | DecimalAI

agent toolingby stijnman

Skill Evolver

stijnman/skill-evolver

Source stijnman/grok-custom-skills ↗License MIT

Full skill rewrite and improvement with versioned backups and templates.

SkillScore57Maintained

Automatic

AI Engineer pack

Last verified 8/28/2026

What it changes

Passed 13 of 23 test cases with the skill, and 9 of 23 without it.

Without skill39%

With skill57%

Checks it passed that the model alone missed

Displays the rollback diff to the user.
Validates the updated skill with hyper-skill-tester.
Validates the update with hyper-skill-tester after approval.

and 3 more in the full report

6 fixed7 passed either way8 still failing2 broke

Pass-rate lift: +17 pp
Tokens: -29%
Agent turns: 0%
Response time: -43%

Model gemini-3.6-flashMeasured 8/28/2026v313 of 23 cases passed with skill

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/stijnman-skill-evolver/SKILL.md https://app.decimal.ai/s/stijnman-skill-evolver/SKILL.md

Want it pinned and updatable? pip install "decimalai>=0.15", then decimalai skills add stijnman-skill-evolver, 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/stijnman-skill-evolver/ and pins it in .decimal/skills.lock

decimalai skills add stijnman-skill-evolver

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

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

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

Safety checks: 3 of 3 passed

Code scan CleanIntent PassedContent Clean

See it in action

One of 6 test cases the skill fixed. Same model, same prompt, with and without the skill.

1 of 6

Prompt

We have just obtained explicit user approval to apply the rewrite diff to our data pipeline SKILL.md. What exact testing tool must be run to validate the newly updated skill?

What was checked, without and with the skill
Checked	Without skill	With skill
Validates the updated skill with hyper-skill-tester.	Missed	Passed

case-09 · gemini-3.6-flash · 8/28/2026 · answers are capped and redacted

What's inside

The instructions your model reads when the skill loads.

SKILL.md38 lines3 filesincludes scripts

Skill Evolver

When to Use

User says evolve skill or task matches this capability
User says upgrade SKILL.md or task matches this capability
User says improve skill file or task matches this capability

Workflow

Create a local versioned backup of the target SKILL.md and read references/evolution-guide.md for the rubric.
Prepare a reviewable rewrite diff that identifies the affected sections, expected benefits, and rollback path.
Obtain explicit user approval before applying, saving, committing, publishing, or otherwise persisting the rewrite.
After approval, apply the rewrite and validate it with hyper-skill-tester. Present any rollback option for approval rather than restoring automatically.

References

Read references/evolution-guide.md when setup, backends, or rubric details are needed.

Integrations

skill-evolution-engine
hyper-skill-tester
natural-language-to-skill

Error Handling

Failure	Response
Broken frontmatter	Preserve the draft, show the local backup and rollback diff, and wait for user approval before restoring.

Gotchas

Read references/evolution-guide.md before major rewrites.

Example

Input: User request matching triggers above. Output: Structured result per workflow with integrations invoked as needed.

Track record

Benchmark

13 / 23

test cases passed, vs 9 / 23 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 — 8/22/2026

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

stijnman / grok-custom-skills

.grok/skills/skill-evolver/SKILL.md

View on GitHub ↗

License MITImported 8/22/2026Version v3Bundled 2 files · scripts

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 Evolver

When to Use

Workflow

References

Integrations

Error Handling

Gotchas

Example

Track record

Benchmark

Adoption

SkillSafety

Source

Other skills for the same job

Related skills