agentleFS
Sign inStart free
I-OnlabsI-Onlabs/claude-code-skills/research-first/SKILL.md

research-first

Skill4 starsChanged 26 days agoMITView on GitHub

When user says any of these, ACTIVATE THIS SKILL: - "enhance", "upgrade", "improve", "optimize" - "fix", "clean up", "simplify" - "make it better", "best practice" - Any config file editing Add to delegation rules:

  • Reads credentials

12 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use council.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • MicrosoftSkill

    council

    Convene the persona panel — six orthogonal review lenses — on a target. Cold independent fan-out, debate-to-consensus, synthesized verdict with recorded dissent and a roster manifest.

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

7 rules. 0 are common, 7 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 10 edited versions we read keep these.

  • 7

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Before deleting or significantly modifying existing code

    line 10
  2. Distinctive:

    Before editing ANY configuration or system file:

    line 15
  3. Distinctive:

    ASK: Do I understand what this does?

    line 17
  4. Distinctive:

    RESEARCH: Use research-oracle agent to find best practices

    line 18
  5. Distinctive:

    VALIDATE: Use code-reviewer to check proposed changes

    line 19
  6. Distinctive:

    When user says any of these, ACTIVATE THIS SKILL:

    line 53
  7. Distinctive:

    Add to delegation rules:

    line 80

“Common” is measured: kept by at least 6 in 10 of the 10 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Research-First Skill

Description

Enforces research and validation before making significant changes to configuration files, system settings, or unfamiliar code.

When This Activates

  • Editing configuration files (settings.json, .env, hooks, etc.)
  • Making changes to unfamiliar systems
  • User asks to "enhance", "upgrade", "fix" something you haven't researched
  • Before deleting or significantly modifying existing code

Methodology

STOP Before Acting

Before editing ANY configuration or system file:

  1. ASK: Do I understand what this does?
  2. RESEARCH: Use research-oracle agent to find best practices
  3. VALIDATE: Use code-reviewer to check proposed changes
  4. CONFIRM: Show user the plan before executing

Required Steps for Config Changes

1. READ current config completely
2. RESEARCH community best practices (use research-oracle)
3. COMPARE current vs best practices
4. PROPOSE changes (show diff to user)
5. WAIT for approval
6. IMPLEMENT incrementally
7. VERIFY with code-reviewer

Agents to Use Proactively

SituationAgentWhy
Unfamiliar systemresearch-oracleLearn before touching
Config changescode-reviewerValidate before/after
Complex editserror-detectiveCatch issues early
Architecturearchitectural-cognition-engineUnderstand impact

Anti-Patterns (DON'T DO)

  • Deleting large sections without understanding why they exist
  • Making "simplifications" that remove functionality
  • Acting on assumptions without verification
  • Editing files before reading them completely
  • Proposing changes without researching alternatives

Trigger Phrases

When user says any of these, ACTIVATE THIS SKILL:

  • "enhance", "upgrade", "improve", "optimize"
  • "fix", "clean up", "simplify"
  • "make it better", "best practice"
  • Any config file editing

Example Workflow

User: "Enhance the settings.local.json"

WRONG:
→ Immediately edit file based on assumptions

RIGHT:
1. Read current file completely
2. Launch research-oracle: "Research Claude Code settings best practices"
3. Wait for research results
4. Compare current config vs recommendations
5. Show user: "Based on research, here are suggested changes..."
6. Wait for approval
7. Make incremental changes
8. Verify with code-reviewer

Integration with CLAUDE.md

Add to delegation rules:

### Before Config/System Changes (→ research-oracle FIRST)
- Any settings.json, settings.local.json edits
- Hook modifications
- System configuration changes
- When user asks to "enhance" or "upgrade"
- **ALWAYS research before modifying**

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free