neuro

name	neuro
description	Mandatory external architectural and code-review harness for non-trivial code changes, PR reviews, refactors, modeling logic, dashboard/API integration, performance-sensitive work, and implementation planning. Must be followed by the code-hardener skill before Gemini edits production code.

Purpose

neuro routes non-trivial engineering work through NEURO / Neurometric before Gemini makes code changes.

Gemini is not the first architectural authority. NEURO provides first-pass architectural review, risk analysis, minimal patch strategy, edge cases, and validation guidance.

NEURO is still advisory. It is not final truth. The current repository, tests, PROJECT_CONTEXT.md, and user instructions remain authoritative.

For non-trivial work, this skill is always step 1 of a two-skill chain:

neuro → code-hardener → Gemini implementation.

Activation

Use this skill for:

• New modules
• Architectural changes
• Cross-file refactors
• API or data-contract changes
• Dashboard/API integration
• Modeling/statistical/quantitative logic
• Performance-sensitive code
• Security-sensitive code
• Code review
• PR review
• Bug fixes that could affect existing behavior
• Any task where the user asks to use NEURO
• Any task where Gemini is uncertain whether a change is trivial

Skip only for trivial changes:

• Typo fixes
• Formatting-only changes
• One-line lint fixes
• Comment cleanup
• Simple unused import removal

If unsure, use NEURO.

Required inputs

Before calling NEURO, gather:

• Original user prompt
• Current PROJECT_CONTEXT.md
• Current branch and git status
• Relevant full file contents
• Neighboring files/contracts
• Existing tests touching the area
• Intended change scope
• Known constraints
• Failure modes to avoid

Required NEURO output

Request NEURO to return:

• Problem summary
• Affected files
• Minimal patch strategy
• Architectural risks
• Edge cases
• Interface/data-contract risks
• Backward compatibility risks
• Tests to run
• Changes to avoid
• Confidence level
• Open questions

Mandatory handoff

After NEURO returns output, Gemini must not edit source code.

Gemini must activate code-hardener and pass the NEURO output into it.

The next required artifact is:

.neuro/chains/latest/04_final_code_hardener_plan.md

No production source file may be edited until that artifact exists and matches the current user request.

Failure behavior

If live NEURO is unavailable:

• Do not fake NEURO output.
• Do not silently use stale cache.
• If the task is risky or architectural, stop and report the blocker.
• If the task is trivial, classify it as trivial and proceed with normal validation.

Progressive disclosure

Keep this SKILL.md concise. Detailed workflow is in:

• workflows/neuro-review-protocol.md

Executable harness is in:

• scripts/neuro_harness.py