Darwin Skill Optimizer
alchaincyf/darwin-skill
Scores SKILL.md files on a nine-dimension rubric, then improves them in a keep-or-revert loop with independent judge agents, test prompts, git history and human checkpoints.
Creates, revises, evaluates and publishes skills in the Open-Science app through its native host.skills composer, with optional test prompts and benchmarks.
Install Open-Science Skill Creator
$ npx skills add aipoch/open-science --skill skill-creator -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install aipoch/open-science skill-creator --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/aipoch/open-science.git skills-src && mkdir -p .claude/skills && cp -r skills-src/resources/skills/skill-creator .claude/skills/skill-creator && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "skill-creator" agent skill from https://github.com/aipoch/open-science/tree/main/resources/skills/skill-creator into .claude/skills/skill-creator/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "skill-creator", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/aipoch/open-science/tree/main/resources/skills/skill-creatorType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add aipoch/open-science --skill skill-creator -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install aipoch/open-science skill-creator --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/aipoch/open-science.git skills-src && mkdir -p .agents/skills && cp -r skills-src/resources/skills/skill-creator .agents/skills/skill-creator && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "skill-creator" agent skill from https://github.com/aipoch/open-science/tree/main/resources/skills/skill-creator into .agents/skills/skill-creator/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "skill-creator", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add aipoch/open-science --skill skill-creator -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install aipoch/open-science skill-creator --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/aipoch/open-science.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/resources/skills/skill-creator .cursor/skills/skill-creator && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "skill-creator" agent skill from https://github.com/aipoch/open-science/tree/main/resources/skills/skill-creator into .cursor/skills/skill-creator/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "skill-creator", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/aipoch/open-science.git --path resources/skills/skill-creator--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add aipoch/open-science --skill skill-creator -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install aipoch/open-science skill-creator --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/aipoch/open-science.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/resources/skills/skill-creator .gemini/skills/skill-creator && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "skill-creator" agent skill from https://github.com/aipoch/open-science/tree/main/resources/skills/skill-creator into .gemini/skills/skill-creator/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "skill-creator", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install aipoch/open-science skill-creatorInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add aipoch/open-science --skill skill-creator -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/aipoch/open-science.git skills-src && mkdir -p .github/skills && cp -r skills-src/resources/skills/skill-creator .github/skills/skill-creator && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "skill-creator" agent skill from https://github.com/aipoch/open-science/tree/main/resources/skills/skill-creator into .github/skills/skill-creator/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "skill-creator", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add aipoch/open-science --skill skill-creator -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install aipoch/open-science skill-creator --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/aipoch/open-science.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/resources/skills/skill-creator .opencode/skills/skill-creator && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "skill-creator" agent skill from https://github.com/aipoch/open-science/tree/main/resources/skills/skill-creator into .opencode/skills/skill-creator/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "skill-creator", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
skill-creator- GitHub stars
- 5.5k
- Token cost
- ~1.7k tokens
- SKILL.md length
- 829 words
- Files
- 17 (incl. scripts, references, assets)
- Skills in repo
- 8
- Repo updated
- First seen
- Licence
- Apache-2.0
At a glance
Creates, revises, evaluates and publishes skills in the Open-Science app through its native host.skills composer, with optional test prompts and benchmarks.
- Works in 6 steps: Capture intent and examples. → Draft or revise the Skill. → Review the package with the user. → …
- Creating a reusable workflow as a skill in the Open-Science app
- SKILL.md covers Native composer, Choose the current stage, Capture intent and Author the package, plus 4 more sections
- Runs JavaScript scripts from its folder
What it does
This skill creates, revises, evaluates and publishes skills for the Open-Science app through its native `host.skills` composer, driven from a JavaScript control-plane REPL. Skills are application-managed packages rather than artifacts. The composer lists, reads, edits, publishes and deletes skills: an edit without `oldString` creates a file and fails if it exists, with `oldString` the old text must occur exactly once, and an existing draft is never silently overwritten. Publishing promotes the whole draft into Personal Skills, and deletion always needs app approval.
The agent infers the current stage and starts there: capture intent and examples, draft or revise, review with you, optionally evaluate realistic prompts, improve from the evidence, and publish only after you accept the draft. Evaluation is not forced, since subjective or exploratory skills may be better reviewed in conversation. Built-in or imported skills are forked into a personal copy under a new lowercase hyphenated name. Bundled helpers include grader, analyzer and comparator agent notes, an eval viewer and scripts to run evals, aggregate benchmarks, validate skills and improve descriptions.
When your agent uses it
- Creating a reusable workflow as a skill in the Open-Science app
- Revising an existing personal skill
- Running test prompts or benchmarks against a skill
- Improving a skill's trigger description
Example prompts
- “Turn my literature screening steps into a reusable skill.”
- “Fork the built-in plotting skill and add a rule for colorblind-safe palettes.”
- “Run a benchmark on my sequence-alignment skill with a few realistic prompts.”
- “My skill never triggers on short requests. Improve its description.”
Requirements
- The Open-Science desktop app, which provides the host.skills composer
Workflow steps
6 steps, taken from the first numbered list in SKILL.md.
- Capture intent and examples.
- Draft or revise the Skill.
- Review the package with the user.
- Optionally evaluate realistic prompts.
- Improve from evidence and repeat.
- Publish only after the user accepts the draft.
What it can do on your machine
Read from SKILL.md and the folder at commit 51d7079. It shows what the files ask for, not the result of running them.
Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
Ships 7 files in scripts/ (JavaScript, from the files we listed), which the agent can run.
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Context cost
Open-Science Skill Creator loads about 1.7k tokens when it runs, and up to ~2.6k if it reads all its reference files. Until then it costs about 68 tokens; SKILL.md has 829 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
Safety
Auto-check passed
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.
SKILL.md
The full file from aipoch/open-science at commit 51d7079, republished under its Apache-2.0 licence (© aipoch). 829 words, ~1,662 tokens.
Download SKILL.mdSave it as
.claude/skills/skill-creator/SKILL.md (or your agent's skills folder). This skill also uses 16 other files; get the full folder from GitHub.- name
- skill-creator
- description
- Create, revise, evaluate, publish, and improve Open-Science Skills through the native JavaScript host.skills composer. Use when the user wants a reusable workflow, an existing Skill changed, test cases or benchmarks for a Skill, or better Skill triggering.
Skill Creator
Create one focused, reusable Skill package. Skills are application-managed packages, not Artifacts.
Use the JavaScript control-plane REPL and the native host.skills composer for lifecycle operations.
Native composer
await host.skills.list()
await host.skills.read(name)
await host.skills.read(name, path)
await host.skills.validate(name)
await host.skills.edit(name, path, content)
await host.skills.edit(name, path, replacement, oldString)
await host.skills.publish(name)
await host.skills.publish(name, true)
await host.skills.delete(stableId)Without oldString, edit creates a file and fails if it exists. With oldString, the old text
must occur exactly once. Never silently overwrite an existing draft file. publish promotes the
complete draft into Personal Skills. delete is privileged and always uses app approval. When a
published Skill and its draft coexist, delete only by the exact draft-<name> or personal-<name>
id returned from list(); never guess from the shared display name.
Choose the current stage
Infer where the user is in the workflow and start there:
- Capture intent and examples.
- Draft or revise the Skill.
- Review the package with the user.
- Optionally evaluate realistic prompts.
- Improve from evidence and repeat.
- Publish only after the user accepts the draft.
Do not force evaluation. Objectively verifiable workflows benefit from test cases; subjective writing or exploratory Skills may be better reviewed directly in conversation.
Capture intent
Extract what is already known from the conversation before asking questions. Confirm only gaps that materially change behavior:
- What should the Skill enable an agent to do?
- When should it trigger, including near-miss cases where it should not?
- What inputs and output formats matter?
- What counts as success, and what failures need explicit handling?
- Are scripts, references, assets, connectors, or example files required?
- Does the user want test cases now?
Prefer one concise question at a time. Calibrate terms such as benchmark, JSON, or assertion to the user's technical comfort.
Author the package
- Call
host.skills.list()before editing. Read every existing file you intend to change. - For Built-in or Imported Skills, create a Personal fork under a new lowercase hyphenated name.
- Use frontmatter with
nameanddescription, plus optionaldisplayName;nameis the immutable safe draft name and defaults as the presentation label whendisplayNameis omitted. - Put stable procedures in
SKILL.md, detailed knowledge inreferences/, deterministic automation inscripts/, and output templates inassets/. - Prefer imperative instructions and explain why constraints matter. Avoid brittle lists of MUSTs.
- Keep
SKILL.mdfocused. Link directly to optional resources and state when to read them. - Re-read changed files, call
host.skills.validate(name), and show the user the important behavior and boundaries before publishing.
Do not promise automatic kernel sidecars, per-Specialist environments, or connector tool patterns; those capabilities are not part of the current composer.
Create test cases
When the user wants evaluation, propose two or three realistic prompts. Ask them to confirm or revise
the set before running anything. Store output-evaluation cases as evals/evals.json. Store trigger
and near-miss cases as trigger-evals.json. Follow references/schemas.md.
Good cases cover different phrasings, input shapes, edge cases, and near misses. Expectations should be observable from the transcript or output files. Use human review for qualities that cannot be reliably reduced to assertions.
Show full SKILL.md (344 more words)Show less
Run and evaluate
Evaluation is capability-gated. First check whether this runtime exposes host.skills.evals. If it
does not, run a qualitative sanity check in the current conversation or publish without evaluation if
the user chooses; never claim that baseline, blind, or trigger evaluation ran when it did not.
When host.skills.evals is available:
- Freeze the draft identity and revision.
- Create paired runs for each case: one with the Skill and one baseline.
- Keep inputs, provider/model, and tool capabilities equal across the pair.
- Save outputs, transcript, timing, token counts, and actual Skill activity.
- Grade expectations using
agents/grader.md. - Aggregate results with
scripts/aggregate-benchmark.js. - Generate the review page with
eval-viewer/generate-review.jsand let the user review outputs before changing the Skill. - Use
agents/comparator.mdonly when A/B origins are genuinely hidden. - Use
agents/analyzer.mdto explain benchmark patterns and comparison results.
Never use persistent host.agents Specialists as pretend isolated evaluators. Never start another
provider CLI from the REPL to bypass the app-owned Session and approval boundaries.
Improve from feedback
Read user feedback, grades, transcripts, and benchmark notes together. Generalize from repeated failures instead of overfitting to one prompt. Look for:
- ambiguous instructions that led to divergent behavior;
- repeated helper code that belongs in
scripts/; - expectations that pass both configurations and therefore do not measure Skill value;
- flaky cases with high variance;
- time or token costs that outweigh the quality gain;
- false-positive and false-negative trigger cases.
scripts/improve-description.js can build and parse a description-improvement prompt, but the current
Agent or an app-owned evaluation Session must perform the model call. Always show description changes
and scores to the user before applying an exact-match edit.
Review and publish
Summarize the final behavior, boundaries, files, and any unverified capability. Publish with
await host.skills.publish(name). Use overwrite = true only after the user explicitly chooses to
replace an existing Personal Skill. Read the published SKILL.md back and report its actual id and
origin.
If the user asks to attach it to a Specialist, read the live Specialist and Skill catalogs first,
then call host.agents.attachSkill(...) and report the returned read-back. Never attach automatically.
© aipoch, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Files
SKILL.md and 16 other files (scripts, references, assets) in resources/skills/skill-creator of aipoch/open-science.
- SKILL.md
- agents/analyzer.md
- agents/comparator.md
- agents/grader.md
- assets/eval_review.html
- eval-viewer/generate-review.js
- eval-viewer/viewer.html
- package.json
- references/schemas.md
- scripts/aggregate-benchmark.js
- scripts/generate-report.js
- scripts/improve-description.js
- scripts/index.js
- scripts/quick-validate.js
- scripts/run-eval.js
- scripts/run-loop.js
- … and 1 more
Open the folder on GitHubat commit 51d7079
Compare with similar skills
Open-Science Skill Creator next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Open-Science Skill Creator this skillaipoch/open-science | 5.5k | — | ~1.7k | Automated safety check: Pass | Apache-2.0 | |
| Darwin Skill Optimizeralchaincyf/darwin-skill | 6.2k | 1 repos | ~4.7k | Automated safety check: Pass | MIT | |
| Skill Quality ReviewerGalaxy-Dawn/claude-scholar | 5.7k | 1 repos | ~3k | Automated safety check: Pass | MIT | |
| OpenCode Skill Creatorantongulin/opencode-skill-creator | 172 | — | ~8.1k | Automated safety check: Pass | Apache-2.0 | |
| Skill Release Gaterohitg00/ai-engineering-from-scratch | 66k | — | ~1k | Automated safety check: Pass | MIT | |
| Skill Creatorzhayujie/CowAgent | 47k | — | ~4.7k | Automated safety check: Notes | MIT |
Similar skills
Skill Quality Reviewer
Galaxy-Dawn/claude-scholar
Scores a skill across description, content organization, writing style and structure, then produces letter grades and a prioritized improvement plan.
OpenCode Skill Creator
antongulin/opencode-skill-creator
Walks you through drafting, testing, evaluating and tuning a skill for OpenCode, from an intake interview to description optimization.
Skill Release Gate
rohitg00/ai-engineering-from-scratch
Evaluates an Agent Skill bundle before release for structure, trigger quality, artifact improvement, script correctness, safety, installed-tree integrity and host portability.
Skill Creator
zhayujie/CowAgent
Guides creating, installing and updating agent skills in a workspace: SKILL.md frontmatter, bundled scripts and references, with scripts to scaffold, validate and package.
Skill Judge
shareAI-lab/Kode-CLI
Evaluates the design quality of an agent skill against official specifications and patterns from existing examples, scoring it and suggesting improvements.
More from aipoch/open-science
All 8 skills in this repoCompute Environment Setup
aipoch/open-science
Prepares setup instructions and a named activation file for a user-managed software environment on an Open-Science SSH or Slurm compute host.
Customize Specialists and Skills
aipoch/open-science
Handles /Customize requests by sending Skill work to the internal skill-creator and managing Specialist agents through the JavaScript host.agents SDK.
Figure Style
aipoch/open-science
Publication-grade correctness and legibility rules for final-deliverable scientific figures, not exploratory plots.
Paper Narrative
aipoch/open-science
Judge and reshape the story told by an entire paper figure deck.
Open-Science Host Inspection
aipoch/open-science
Teaches an agent to inspect Open-Science's JavaScript control REPL, check which host.* calls are allowed, and find project files, sessions and agent frames.
Figure Composer
aipoch/open-science
Compose one publication-grade multi-panel figure. An agent skill from aipoch/open-science.
Related
Works with
Categories
Questions about Open-Science Skill Creator
What does Open-Science Skill Creator do?
Creates, revises, evaluates and publishes skills in the Open-Science app through its native host.skills composer, with optional test prompts and benchmarks. skills` composer, driven from a JavaScript control-plane REPL. Skills are application-managed packages rather than artifacts.
When should I use Open-Science Skill Creator?
Open-Science Skill Creator fits situations like: creating a reusable workflow as a skill in the Open-Science app; revising an existing personal skill; running test prompts or benchmarks against a skill; improving a skill's trigger description.
How do I install Open-Science Skill Creator in Claude Code?
Run `npx skills add aipoch/open-science --skill skill-creator -a claude-code`. Or copy the skill folder (resources/skills/skill-creator in aipoch/open-science) into .claude/skills/skill-creator in your project. Claude Code loads it when a task matches its description.
How do I install Open-Science Skill Creator in Codex?
Run `npx skills add aipoch/open-science --skill skill-creator -a codex`. Or copy the skill folder (resources/skills/skill-creator in aipoch/open-science) into .agents/skills/skill-creator in your project. Codex loads it when a task matches its description.
Can I use Open-Science Skill Creator in Cursor, Gemini CLI or GitHub Copilot?
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add aipoch/open-science --skill skill-creator -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/skill-creator, .gemini/skills/skill-creator, .github/skills/skill-creator and .opencode/skills/skill-creator in your project.
What does Open-Science Skill Creator need to run?
Going by SKILL.md and its folder, Open-Science Skill Creator needs JavaScript for the scripts in its folder. Our summary lists: The Open-Science desktop app, which provides the host.skills composer.
Does Open-Science Skill Creator access the network?
SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.
Is Open-Science Skill Creator safe to install?
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.
What licence does Open-Science Skill Creator use?
Open-Science Skill Creator is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does Open-Science Skill Creator use?
About 1.7k tokens (SKILL.md is roughly 6.6k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 956 tokens, read only when the agent opens those files.
What are the alternatives to Open-Science Skill Creator?
Skills that share tags, products or a category with Open-Science Skill Creator: Darwin Skill Optimizer (alchaincyf/darwin-skill, 6.2k stars), Skill Quality Reviewer (Galaxy-Dawn/claude-scholar, 5.7k stars), OpenCode Skill Creator (antongulin/opencode-skill-creator, 172 stars) and Skill Release Gate (rohitg00/ai-engineering-from-scratch, 66k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Open-Science Skill Creator?
aipoch (a GitHub organization) maintains it in aipoch/open-science, which has 5,475 GitHub stars. The repository holds 8 skills in this directory. The repository was last updated on October 8, 2026.
Source: aipoch/open-science on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.