feat: update research skills part 4

[sixiang-svg]

New Skill Contribution

Skill Info

• Skill ID:
• Category:
• Description:

Checklist

• SKILL.md has valid YAML frontmatter with all required fields
• Skill ID matches the directory name
• Category is one of: productivity, development, communication, writing, research, other
• Instructions are clear and self-contained
• Instructions are under 4000 tokens
• No external URLs referenced in instructions
• Tested the skill with an AI agent and it produces good results
• Added usage examples in the frontmatter

Testing

Describe how you tested this skill:

Notes

Any additional context or notes for reviewers.

[greptile-apps]

Greptile Summary

This PR continues a multi-part effort to standardize Research skill files by: adding requires: [] and examples: to frontmatter, improving description fields, and replacing thin placeholder "When to Use" sections with substantive Instruction, When to Use, and Output sections. Two previously mismatched skills (gget — Biopython content under a gget ID — and gwas-database — Biomni content under a gwas-database ID) are deleted, which is a positive cleanup.

Key observations:

• Deletions are correct — removing gget and gwas-database eliminates the content-to-ID mismatches flagged in the previous review cycle.
• skills/Research/gtars/SKILL.md — All three Common Workflow sections and the Installation subsections retain their introductory sentences (e.g. "Install gtars Python bindings:", "Identify overlapping genomic features:") but their code blocks were stripped, leaving orphaned dangling text with no actionable content.
• skills/Research/fluidsim/SKILL.md — The new Instruction section refers to the "AttributeError feedback system," but AttributeError is a Python exception, not a feedback system. The terminology is inaccurate and may confuse users.
• Several issues from the previous review cycle persist and are not resolved in this PR: empty ## Installation in geniml, empty Core Capabilities subsections in fluidsim, missing trailing newlines across 11+ files, non-standard version: field and removed allowed-tools in hil-test-automation, and removed allowed-tools in geo-fundamentals.

Confidence Score: 3/5

• Safe to merge with caveats — documentation-only changes carry no runtime risk, but several skills still have incomplete or inaccurate content that reduces their usefulness.
• The PR makes genuine improvements (better descriptions, substantive Instruction sections, correct deletions of mismatched skills), but multiple issues from the previous review cycle remain unaddressed (empty sections, missing trailing newlines, removed allowed-tools), and new issues were introduced (orphaned dangling text in gtars, inaccurate "AttributeError feedback system" terminology in fluidsim).
• skills/Research/gtars/SKILL.md and skills/Research/fluidsim/SKILL.md need the most attention due to incomplete content sections and inaccurate terminology respectively.

Important Files Changed

Filename	Overview
skills/Research/fda-database/SKILL.md	All FDAQuery Python code examples, Common Query Patterns, Working with Results, and Best Practices sections removed; replaced with high-level text instructions. Loses programmatic usage guidance but gains a cleaner instruction set.
skills/Research/fluidsim/SKILL.md	Frontmatter updated and Instruction section added, but Core Capabilities subsections still carry orphaned headings without content (previously flagged), and new "AttributeError feedback system" terminology is technically inaccurate.
skills/Research/geniml/SKILL.md	Instruction/Output sections added with good content; Installation section heading still empty (previously flagged); double uv uv pip install commands correctly removed.
skills/Research/geo-fundamentals/SKILL.md	Description clarified to reflect GEO = Generative Engine Optimization; Instruction/When to Use/Output added; allowed-tools restriction removed without justification (previously flagged).
skills/Research/gget/SKILL.md	File deleted — resolves previously flagged content-to-ID mismatch (Biopython content under gget ID).
skills/Research/gpt-researcher/SKILL.md	Quick Start Python examples, architecture diagram, and configuration reference removed; replaced with text-only Instruction section; missing trailing newline (previously flagged).
skills/Research/gtars/SKILL.md	Installation subsections and all three Common Workflow sections retain introductory text but have their code blocks stripped, leaving orphaned dangling sentences with no actionable content.
skills/Research/gwas-database/SKILL.md	File deleted — resolves previously flagged content-to-ID mismatch (Biomni content under gwas-database ID).
skills/Research/hil-test-automation/SKILL.md	Output section added; version field promoted to top-level frontmatter (non-standard) and allowed-tools removed (both previously flagged).

Flowchart

%%{init: {'theme': 'neutral'}}%%
flowchart TD
    PR["PR #56: Research Skills Update (Part 4)"]

    PR --> DEL["Deletions (content-mismatch fixes)"]
    PR --> UPD["Updates (18 files)"]

    DEL --> D1["gget/SKILL.md ✅\n(Biopython content removed)"]
    DEL --> D2["gwas-database/SKILL.md ✅\n(Biomni content removed)"]

    UPD --> FM["Frontmatter Standardization\nrequires: []\nexamples: [...]"]
    UPD --> SECT["Section Restructuring\nInstruction / When to Use / Output"]
    UPD --> STRIP["Code Block Removal"]

    STRIP --> S1["fluidsim ⚠️\nCore Capabilities empty headings\nAttributeError 'feedback system'"]
    STRIP --> S2["gtars ⚠️\nInstallation + Workflow sections\nhave orphaned intro text, no code"]
    STRIP --> S3["geniml ⚠️\nInstallation heading still empty"]
    STRIP --> S4["gpt-researcher ⚠️\nQuick Start + architecture removed"]

    FM --> PERSIST["Persisting Issues"]
    SECT --> PERSIST
    PERSIST --> P1["Missing trailing newlines\n(11+ files)"]
    PERSIST --> P2["hil-test-automation:\nversion top-level, no allowed-tools"]
    PERSIST --> P3["geo-fundamentals:\nallowed-tools removed"]

Loading

_{Last reviewed commit: "Delete: body mismatc..."}