intent-detection

Intent Detection -- Workflow Routing

When user describes work WITHOUT specifying a /rx: command, analyze their intent and suggest the appropriate workflow BEFORE starting work.

Routing Table

Signal	Detected Intent	Suggest
"bug", "error", "crash", "failing", "broken", stack trace	Bug investigation	/rx:investigate
"add", "implement", "build", "create" + multi-step	New feature	/rx:plan
"review", "check", "audit" code	Code review	/rx:review
"fix" + small/specific scope	Quick fix	handle directly or /rx:quick
"refactor", "clean up", "improve"	Refactoring	/rx:plan (needs scope)
"research", "how to", "what's the best"	Research	/rx:research
"evaluate", "compare", "library", "should we use"	Library evaluation	/rx:research
"test", "spec", "coverage"	Testing	handle directly or /rx:plan
Describes 1-2 file changes, < 50 lines	Small task	handle directly
"performance", "slow", "re-render", "bundle size"	Performance	/rx:rerender-check or /rx:audit
"design", "tokens", "spacing", "colors", "theme"	Design system	/rx:design
"trace", "render tree", "prop flow", "why re-renders"	Component tracing	/rx:trace
"circular", "import", "dead code", "coupling"	Boundary analysis	/rx:boundaries
"tech debt", "TODO", "outdated", "deprecated"	Tech debt	/rx:techdebt
"document", "jsdoc", "readme", "adr"	Documentation	/rx:document
"challenge", "stress test", "edge case"	Adversarial review	/rx:challenge
"that worked", "fixed it", "problem solved"	Knowledge capture	/rx:compound
"enhance plan", "more detail", "deepen"	Plan enhancement	/rx:plan --existing
"triage", "which findings", "prioritize fixes"	Finding triage	/rx:triage
"props audit", "state location", "effect deps"	Props audit	/rx:props-audit
"deploy", "release", "production"	Pre-deploy	/rx:verify

Behavior

1. Read user's first message
2. Match against routing table (use keyword + context signals, not exact match)
3. If match found with multi-step workflow: "This looks like [intent]. I'd suggest [command] -- want me to run it, or should I just dive in?"
4. If trivial task (typo, single-line fix, config change): skip suggestion, just do it
5. If user already specified a /rx: command: follow it, do not re-suggest
6. NEVER block the user -- suggestion only, not mandatory

Confidence Signals

High confidence (suggest immediately):

• Stack trace or error message pasted -> /rx:investigate
• "Add [feature] with [multiple components]" -> /rx:plan
• "Review my changes" or "check this code" -> /rx:review
• "Why does [component] re-render" -> /rx:trace
• "Set up design tokens" or "initialize design system" -> /rx:design init

Medium confidence (suggest with caveat):

• "Fix [thing]" -- could be quick or complex, suggest based on scope description
• "Update [thing]" -- could be small edit or refactor
• "Improve performance" -- could be /rx:rerender-check or /rx:audit

Low confidence (just do it):

• Single file mentioned, clear change
• "Change X to Y"
• Configuration or dependency updates

Complexity Signals

When a task matches a workflow command, check complexity before suggesting:

Trivial signals (suggest /rx:quick or handle directly):

• Single file mentioned explicitly
• "exclude X from Y", "add X to config", "rename", "change X to Y"
• Problem + solution both stated ("X is wrong, change to Y")
• One-line fix described

Complex signals (suggest /rx:plan or /rx:investigate):

• 3+ modules or files mentioned
• "intermittent", "race condition", "sometimes", "random"
• Stack trace with 5+ frames
• "across", "all", "every" (scope indicators)
• Multiple state domains involved (auth + API + UI)

Override rule: If user invokes /rx:full but task matches trivial signals: "This looks like a quick fix. Want /rx:quick instead, or stick with the full cycle?"

Iron Laws

1. NEVER block on suggestion -- If user starts explaining, just do the work
2. One suggestion max -- Do not re-suggest if user ignores first suggestion
3. Commands are shortcuts, not gates -- All work can be done without commands

Integration

This skill is consulted at session start. It works alongside:

• SessionStart hooks (show plugin loaded message, check for resumable work)
• CLAUDE.md routing instructions (passive reference)
• Individual workflow skills (activated by commands)