unity-audit
Full project health check — meta file integrity, missing references, assembly definition graph, code quality scan, scene hierarchy audit.
Manage the project's instinct library — view, promote, evolve, export, or import atomic learned behaviors captured automatically by PostToolUse observations.
> /plugin marketplace add XeldarAlz/everything-claude-unityHow it fires
How this command gets triggered: by you, by Claude, or both.
/unity-instinctsContext preview
What this command does when you run it.
Manage the project's instinct library — view, promote, evolve, export, or import atomic learned behaviors captured automatically by PostToolUse observations.
name: unity-instincts description: "Manage the project's instinct library — view, promote, evolve, export, or import atomic learned behaviors captured automatically by PostToolUse observations." user-invocable: true args: subcommand
Manage atomic Unity-specific learnings: **$ARGUMENTS**
Instincts are atomic learned behaviors captured by `instinct-capture.sh` (PostToolUse) and distilled by `instinct-distill.sh` (Stop). Each instinct has: a trigger, an action, a confidence score (0.3–0.9), an evidence count, and a scope (project or global).
Storage:
Show a dashboard of the instinct library for the current project.
1. Compute project hash: `git config --get remote.origin.url | shasum | head -c 12` (or fallback to `git rev-parse --show-toplevel`). 2. List files under `.claude/state/instincts/project/<hash>/` and `.claude/state/instincts/global/`. 3. For each instinct, read JSON fields and render:
## Instinct Library **Project:** <project-hash> **Project instincts:** <count> (<count-above-0.7> high-confidence) **Global instincts:** <count> **Raw observations:** <count> (this session: <count>) ### High confidence (>= 0.7) — project | Trigger | Action | Confidence | Evidence | |---|---|---|---| | before editing *View.cs | expect quality-gate warnings; read Model first | 0.8 | 12 | | ... | ... | ... | ... | ### Medium (0.5–0.69) — project ... ### Low (0.3–0.49) — project ... ### Global (all scopes) ...
Sort within each band by `evidence_count` desc.
Filter and list instincts. Useful when the library grows:
/unity-instincts list --domain mvs /unity-instincts list --min-confidence 0.7
Read JSON files, filter, output the same table format as `status` but limited to the filtered set.
Graduate high-confidence instincts into draft skill files.
1. Collect instincts from project and global scopes with `confidence >= N` (default 0.8) AND `evidence_count >= 5`. 2. Group by `domain`. 3. For each domain cluster, draft a SKILL.md:
--- name: learned-<domain>-patterns description: "Patterns auto-learned from Unity sessions in this project" alwaysApply: false globs: ["Assets/Scripts/**/*.cs"] --- # Learned <Domain> Patterns ## Trigger <trigger from highest-confidence instinct> ## Patterns - <trigger>: <action> (seen <n> times) - ... ## Source Evolved from instincts: <list of instinct ids>
4. Output the draft to stdout; do NOT write it. Tell the user: > Draft skill at `.claude/skills/core/learned-<domain>-patterns/SKILL.md` — review before installing.
Move a project instinct to the global scope. Normally only done when the same trigger/action has been seen in 2+ projects, but `--force` bypasses the check.
1. Read `.claude/state/instincts/project/<project-hash>/<instinct-id>.json`. 2. Without `--force`: look for an identical trigger in any OTHER project hash. If none found, refuse and say so. 3. Rewrite `scope: "global"`, drop `project_id`, write to `.claude/state/instincts/global/<instinct-id>.json`. 4. Delete the project copy.
Inverse of promote. Moves a global instinct back to the current project, typically when it turned out to be project-specific.
Export all instincts (project + global) as a single JSON file. Default output: `.claude/state/instincts/export-<date>.json`.
Import instincts from a file produced by `export`. Duplicate IDs are merged: `evidence_count` sums, `confidence` is the max, `last_seen` is the max.
Delete state. Requires explicit flag — no accidental wipes.
Ask for confirmation before running.
The ultimate Claude Code toolkit for Unity game development. A production-ready, plug-and-play system that gives Claude Code deep Unity expertise — from writing performant C# to building scenes, profiling performance, and triggering iOS/Android builds — all
Full project health check — meta file integrity, missing references, assembly definition graph, code quality scan, scene hierarchy audit.
Configures and triggers Unity builds via MCP — handles platform settings, scenes, player settings, and monitors build progress.
Diagnostic health check — verifies MCP connectivity, .claude/ integrity, Unity project structure, and hook registration.
Plans and implements a Unity feature — identifies subsystems, loads skills, writes code, sets up scene elements via MCP.
Diagnoses and fixes a Unity bug — reads console errors, checks common causes, applies targeted fix, verifies via MCP.
Scans a Unity project and generates a tailored CLAUDE.md with detected configuration, packages, render pipeline, and recommended skills.