sci-tools

스타8

포크0

업데이트2026년 4월 26일 20:03

Browse and search the Harvard ToolUniverse catalog of 2,200+ biomedical tools from the CLI, and create custom scientific skills with validated templates. Two modes: browse (search, filter, drill-down on biomedical tools) and create (generate sci-* skills from natural language descriptions with scientific defaults). Triggers on: "tools for", "find tools", "search tools", "browse tools", "ToolUniverse", "biomedical tools", "what tools exist for", "tool catalog", "refresh tools", "update catalog", "create a research skill", "build a science skill", "new scientific skill", "custom research skill". Does NOT trigger for: general "create a skill" without science context (use meta-skill-creator), data analysis (use sci-data-analysis), literature search (use sci-literature-research).

설치

Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.

출처

krmdel

krmdel/organon

GitHub 저장소 열기 Creator 저장소 보기

다운로드

Tool Ecosystem

Outcome

Browse 2,200+ biomedical tools from Harvard's ToolUniverse catalog with instant local search, or create custom sci-* skills from natural language descriptions. Browse results display as compact tables (per D-03). Custom skills are validated and auto-registered into the ecosystem.

Context Needs

File	Load level	Purpose
`research_context/research-profile.md`	full	Field context for skill creation personalization
`context/learnings.md`	`## sci-tools` section	Previous feedback

Dependencies

Skill	Required?	What it provides	Without it
`meta-skill-creator`	Required (for create mode)	Full skill creation workflow	Browse mode still works; create mode unavailable

Step 0: Auto-Setup

Run .claude/skills/sci-tools/scripts/setup.sh if first invocation. Downloads the ToolUniverse catalog snapshot to data/tooluniverse-catalog.json if not present.

Step 1: Detect Intent

Parse user request into mode:

browse -- "tools for", "find tools", "search tools", "browse tools", "ToolUniverse", "biomedical tools", "what tools exist for", "tool catalog", "show categories", "list categories"
refresh -- "refresh tools", "update catalog", "re-download catalog"
create -- "create a research skill", "build a science skill", "new scientific skill", "custom research skill", "I need a skill that" (in science context)

If ambiguous, ask which mode.

Step 2: Browse Mode (TOOL-01, TOOL-04)

Run catalog search using catalog_ops.search_catalog(query, category=category, limit=20).
If user specified a category, use it. If category doesn't match exactly, use fuzzy matching (substring). If no results, show available categories via catalog_ops.list_categories().
Display results using catalog_ops.format_results_table(results) -- compact table with Name, Category, Type, Description (per D-03).
Show result count and "last refreshed" date from catalog metadata.
Ask: "Want details on any of these tools? Name one to drill down."

Step 3: Drill-Down (TOOL-04)

Run catalog_ops.get_tool_details(tool_name) to get full schema via tu info --json.
Display: full description, parameters (name, type, required), return schema, test examples, usage notes.
If MCP tools available (mcp__tooluniverse__get_tool_info), use those as alternative for richer documentation.

Step 4: Refresh Mode (TOOL-01)

Run catalog_ops.refresh_catalog().
Report: "Catalog refreshed: {total_tools} tools. Last update: {refreshed_at}."
If refresh fails (network error, uvx unavailable), report error and note catalog still usable from last snapshot.

Step 5: Create Mode (TOOL-03, TOOL-05)

Load research_context/research-profile.md for field context.
Get scientific defaults from validate_ops.get_scientific_defaults():
- Pre-fill category prefix: sci-
- Pre-fill output path: projects/sci-{skill-name}/
- Pre-fill context needs: research-profile.md (full)
- Include reproducibility logging step
Read .claude/skills/sci-tools/references/sci-skill-template.md for the template structure.
Read .claude/skills/meta-skill-creator/SKILL.md and follow its methodology starting from "Capture Intent", but with these pre-filled values as defaults. The scientist only needs to describe what the skill should do -- scientific conventions are assumed (per D-06, D-07).
After scaffold is generated: offer to test (per D-08). Scientist can say "looks good" to skip eval, or provide test prompts to iterate.

Step 6: Validate Created Skill (TOOL-05)

Run validate_ops.validate_skill(skill_dir). Report errors and warnings.
Run validate_ops.check_trigger_conflicts(skill_dir). If conflicts found, show them with suggestions (per D-11): "Trigger phrase '{phrase}' conflicts with {existing_skill}. Suggested alternative: '{suggestion}'." Non-blocking -- scientist decides.
If no errors: skill is ready. Heartbeat reconciliation will auto-register it in CLAUDE.md on next session (per D-10).
If errors: show them and offer to fix.

Step 7: Feedback

Ask: "How did this land? Any adjustments?" Log feedback to context/learnings.md under ## sci-tools.

Rules

Always run setup.sh before first catalog search
Always show compact table format for browse results (per D-03)
Always use local catalog for search (per D-01), MCP/CLI for drill-down only (per D-02)
Never duplicate meta-skill-creator logic -- delegate to it (per D-06)
Always pre-fill scientific defaults for create mode (per D-07)
Always run validation after skill creation (per D-09)
Trigger conflicts are warnings, not blockers (per D-11)
SkillsMP marketplace is deferred to v2 (per D-05). If asked, explain it's planned for a future version.

Self-Update

If the user flags an issue, update the ## Rules section with the correction and today's date.

이 저장소의 다른 Skills

같은 저장소

sci-writing

krmdel/organon

Draft manuscript sections, format citations, and get AI peer review. Natural language invocation -- describe what you need and the skill routes to drafting, citation formatting, or peer review. Reads research profile, data analysis outputs, and literature search results for context-aware writing. After drafting, offers to generate supporting figures via viz skills. Triggers on: "draft introduction", "write methods", "format citations", "APA style", "bibliography", "peer review", "review my draft", "write abstract". Does NOT trigger for: literature search (use sci-literature-research), data analysis (use sci-data-analysis), image generation (use viz-nano-banana), repurpose/blog/lay summary (use sci-communication).

2026-07-028

viz-figure-mirror

krmdel/organon

Transfer the visual style of a reference paper figure onto the user's own data. Takes a screenshot of a figure from a paper plus tabular data and produces an editable matplotlib script in that figure's register, gated by a mechanical quality floor and a Drawer/Reviewer refinement loop. Output is a self-contained .py + PNG + Type-42 PDF. Triggers on: "match this figure style", "figure in the style of", "make my plot look like this paper", "publication figure from reference", "style transfer figure", "reproduce this figure with my data", "NeurIPS-quality version of this chart". Does NOT trigger for: plain data plots with no reference image (use sci-data-analysis), AI-generated illustrations (use viz-nano-banana), flowcharts/architecture/sequence diagrams (use viz-diagram-code), hand-drawn sketches (use viz-excalidraw-diagram).

2026-07-028

tool-substack

krmdel/organon

Publish and edit markdown blog posts on Substack as drafts — converts markdown to Substack's ProseMirror schema, uploads local images to Substack's CDN, pre-renders mermaid diagrams to PNG, and creates or updates drafts via Substack's private API. Draft-only by design — publish is always a human click. Triggers on: "push to substack", "substack draft", "publish to substack", "create substack draft", "send to substack", "substack this post", "update substack", "edit substack draft", "list substack drafts". Does NOT trigger for: reading existing Substack posts, subscriber management, newsletter scheduling. Does NOT publish — only creates/updates drafts you review and publish by hand.

2026-05-128

meta-skill-creator

krmdel/organon

Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.

2026-04-268

meta-wrap-up

krmdel/organon

End-of-session checklist that reviews deliverables, collects feedback, fixes skills, updates learnings, and commits work. Use when the user says "wrap up", "close session", "end session", "wrap things up", "we're done", "that's it for today", "session done", or invokes /wrap-up. Run at the end of any working session or after completing a major deliverable. Does NOT trigger for content writing, voice extraction, positioning, or audience research.

2026-04-268

ops-cron

krmdel/organon

Schedule recurring tasks that run automatically via a system-level watchdog, even when Claude Code is closed. Job definitions persist in cron/jobs/ as the registry. The watchdog (launchd on Mac, Task Scheduler on Windows) checks every hour and runs due jobs headlessly via claude -p. Triggers on: "schedule a job", "cron job", "run this every morning", "automate daily", "recurring task", "set up a scheduled job", "check scheduled jobs", "list scheduled jobs", "install watchdog", "uninstall watchdog". Does NOT trigger for one-off tasks or in-session reminders.

2026-04-268

name

sci-tools

description

Tool Ecosystem

Outcome

Context Needs

File	Load level	Purpose
`research_context/research-profile.md`	full	Field context for skill creation personalization
`context/learnings.md`	`## sci-tools` section	Previous feedback

Dependencies

Skill	Required?	What it provides	Without it
`meta-skill-creator`	Required (for create mode)	Full skill creation workflow	Browse mode still works; create mode unavailable

Step 0: Auto-Setup

Run .claude/skills/sci-tools/scripts/setup.sh if first invocation. Downloads the ToolUniverse catalog snapshot to data/tooluniverse-catalog.json if not present.

Step 1: Detect Intent

Parse user request into mode:

browse -- "tools for", "find tools", "search tools", "browse tools", "ToolUniverse", "biomedical tools", "what tools exist for", "tool catalog", "show categories", "list categories"
refresh -- "refresh tools", "update catalog", "re-download catalog"
create -- "create a research skill", "build a science skill", "new scientific skill", "custom research skill", "I need a skill that" (in science context)

If ambiguous, ask which mode.

Step 2: Browse Mode (TOOL-01, TOOL-04)

Run catalog search using catalog_ops.search_catalog(query, category=category, limit=20).
If user specified a category, use it. If category doesn't match exactly, use fuzzy matching (substring). If no results, show available categories via catalog_ops.list_categories().
Display results using catalog_ops.format_results_table(results) -- compact table with Name, Category, Type, Description (per D-03).
Show result count and "last refreshed" date from catalog metadata.
Ask: "Want details on any of these tools? Name one to drill down."

Step 3: Drill-Down (TOOL-04)

Run catalog_ops.get_tool_details(tool_name) to get full schema via tu info --json.
Display: full description, parameters (name, type, required), return schema, test examples, usage notes.
If MCP tools available (mcp__tooluniverse__get_tool_info), use those as alternative for richer documentation.

Step 4: Refresh Mode (TOOL-01)

Run catalog_ops.refresh_catalog().
Report: "Catalog refreshed: {total_tools} tools. Last update: {refreshed_at}."
If refresh fails (network error, uvx unavailable), report error and note catalog still usable from last snapshot.

Step 5: Create Mode (TOOL-03, TOOL-05)

Load research_context/research-profile.md for field context.
Get scientific defaults from validate_ops.get_scientific_defaults():
- Pre-fill category prefix: sci-
- Pre-fill output path: projects/sci-{skill-name}/
- Pre-fill context needs: research-profile.md (full)
- Include reproducibility logging step
Read .claude/skills/sci-tools/references/sci-skill-template.md for the template structure.
Read .claude/skills/meta-skill-creator/SKILL.md and follow its methodology starting from "Capture Intent", but with these pre-filled values as defaults. The scientist only needs to describe what the skill should do -- scientific conventions are assumed (per D-06, D-07).
After scaffold is generated: offer to test (per D-08). Scientist can say "looks good" to skip eval, or provide test prompts to iterate.

Step 6: Validate Created Skill (TOOL-05)

Run validate_ops.validate_skill(skill_dir). Report errors and warnings.
Run validate_ops.check_trigger_conflicts(skill_dir). If conflicts found, show them with suggestions (per D-11): "Trigger phrase '{phrase}' conflicts with {existing_skill}. Suggested alternative: '{suggestion}'." Non-blocking -- scientist decides.
If no errors: skill is ready. Heartbeat reconciliation will auto-register it in CLAUDE.md on next session (per D-10).
If errors: show them and offer to fix.

Step 7: Feedback

Ask: "How did this land? Any adjustments?" Log feedback to context/learnings.md under ## sci-tools.

Rules

Always run setup.sh before first catalog search
Always show compact table format for browse results (per D-03)
Always use local catalog for search (per D-01), MCP/CLI for drill-down only (per D-02)
Never duplicate meta-skill-creator logic -- delegate to it (per D-06)
Always pre-fill scientific defaults for create mode (per D-07)
Always run validation after skill creation (per D-09)
Trigger conflicts are warnings, not blockers (per D-11)
SkillsMP marketplace is deferred to v2 (per D-05). If asked, explain it's planned for a future version.

Self-Update

If the user flags an issue, update the ## Rules section with the correction and today's date.