dean0x · dean0x · Mar 1, 2026 · Mar 1, 2026
@@ -56,6 +56,13 @@
       "version": "1.0.0",
       "keywords": ["skills", "quality", "patterns"]
     },
+    {
+      "name": "devflow-ambient",
+      "source": "./plugins/devflow-ambient",
+      "description": "Ambient mode — proportional quality enforcement without explicit commands",
+      "version": "1.0.0",
+      "keywords": ["ambient", "routing", "quality", "tdd"]
+    },
     {
       "name": "devflow-audit-claude",
       "source": "./plugins/devflow-audit-claude",

@@ -12,7 +12,7 @@ DevFlow enhances Claude Code with intelligent development workflows. Modificatio
 
 ## Architecture Overview
 
-Plugin marketplace with 8 self-contained plugins, each following the Claude plugins format (`.claude-plugin/plugin.json`, `commands/`, `agents/`, `skills/`).
+Plugin marketplace with 9 self-contained plugins, each following the Claude plugins format (`.claude-plugin/plugin.json`, `commands/`, `agents/`, `skills/`).
 
 | Plugin | Purpose | Teams Variant |
 |--------|---------|---------------|
@@ -22,6 +22,7 @@ Plugin marketplace with 8 self-contained plugins, each following the Claude plug
 | `devflow-resolve` | Review issue resolution | Optional |
 | `devflow-debug` | Competing hypothesis debugging | Optional |
 | `devflow-self-review` | Self-review (Simplifier + Scrutinizer) | No |
+| `devflow-ambient` | Ambient mode — proportional quality enforcement | No |
 | `devflow-core-skills` | Auto-activating quality enforcement | No |
 | `devflow-audit-claude` | Audit CLAUDE.md files (optional) | No |
 
@@ -35,13 +36,13 @@ Commands with Teams Variant ship as `{name}.md` (parallel subagents) and `{name}
 
 ```
 devflow/
-├── shared/skills/          # 24 skills (single source of truth)
+├── shared/skills/          # 26 skills (single source of truth)
 ├── shared/agents/          # 10 shared agents (single source of truth)
-├── plugins/devflow-*/      # 8 self-contained plugins
+├── plugins/devflow-*/      # 9 self-contained plugins
 ├── docs/reference/         # Detailed reference documentation
 ├── scripts/                # Helper scripts (statusline, docs-helpers)
-│   └── hooks/              # Working Memory hooks (stop, session-start, pre-compact)
-├── src/cli/                # TypeScript CLI (init, list, uninstall)
+│   └── hooks/              # Working Memory + ambient hooks (stop, session-start, pre-compact, ambient-prompt)
+├── src/cli/                # TypeScript CLI (init, list, uninstall, ambient)
 └── .claude-plugin/         # Marketplace registry
 ```
 
@@ -92,6 +93,7 @@ All generated docs live under `.docs/` in the project root:
 - `/resolve` — N Resolver agents + Git
 - `/debug` — Agent Teams competing hypotheses
 - `/self-review` — Simplifier then Scrutinizer (sequential)
+- `/ambient` — Intent classification + proportional skill loading (no agents, main session only)
 - `/audit-claude` — CLAUDE.md audit (optional plugin)
 
 **Shared agents** (10): git, synthesizer, skimmer, simplifier, coder, reviewer, resolver, shepherd, scrutinizer, validator

@@ -24,9 +24,9 @@ After setup, DevFlow commands (`/code-review`, `/implement`, etc.) are available
 
 ```
 devflow/
-├── shared/skills/       # 24 skills (single source of truth)
+├── shared/skills/       # 26 skills (single source of truth)
 ├── shared/agents/       # 10 shared agents (single source of truth)
-├── plugins/devflow-*/   # 8 self-contained plugins
+├── plugins/devflow-*/   # 9 self-contained plugins
 ├── scripts/hooks/       # Working Memory hooks
 ├── src/cli/             # TypeScript CLI (init, list, uninstall)
 ├── tests/               # Test suite (Vitest)

@@ -24,7 +24,7 @@ DevFlow adds structured commands that handle the full lifecycle: specify feature
 - **Full-lifecycle implementation** — spec, explore, plan, code, validate, refine in one command
 - **Automatic session memory** — survives restarts, `/clear`, and context compaction
 - **Parallel debugging** — competing hypotheses investigated simultaneously
-- **24 quality skills** — 11 auto-activating, plus specialized review and agent skills
+- **26 quality skills** — 12 auto-activating, plus specialized review and agent skills
 
 ## Quick Start
 
@@ -48,6 +48,7 @@ Then in Claude Code:
 | `devflow-resolve` | `/resolve` | Process review issues — fix or defer to tech debt |
 | `devflow-debug` | `/debug` | Parallel hypothesis debugging |
 | `devflow-self-review` | `/self-review` | Self-review workflow (Simplifier + Scrutinizer) |
+| `devflow-ambient` | `/ambient` | Ambient mode — proportional quality enforcement |
 | `devflow-core-skills` | (auto) | Auto-activating quality enforcement skills |
 
 ## Command Details
@@ -116,6 +117,7 @@ The `devflow-core-skills` plugin provides quality enforcement skills that activa
 | `git-safety` | Rebasing, force-pushing, merge conflicts |
 | `git-workflow` | Staging files, creating commits, PRs |
 | `github-patterns` | GitHub API operations, PR comments, releases |
+| `test-driven-development` | Implementing new features (RED-GREEN-REFACTOR) |
 | `test-patterns` | Writing or modifying tests |
 | `input-validation` | Creating API endpoints |
 | `typescript` | Working in TypeScript codebases |
@@ -209,6 +211,8 @@ Session context is saved and restored automatically via Working Memory hooks —
 | `npx devflow-kit init` | Install all plugins |
 | `npx devflow-kit init --plugin=<names>` | Install specific plugin(s) |
 | `npx devflow-kit list` | List available plugins |
+| `npx devflow-kit ambient --enable` | Enable always-on ambient mode |
+| `npx devflow-kit ambient --disable` | Disable ambient mode |
 | `npx devflow-kit uninstall` | Remove DevFlow |
 
 ### Init Options

@@ -0,0 +1,7 @@
+{
+  "name": "devflow-ambient",
+  "description": "Ambient mode — proportional quality enforcement without explicit commands",
+  "version": "1.0.0",
+  "agents": [],
+  "skills": ["ambient-router"]
+}
@@ -0,0 +1,49 @@
+# devflow-ambient
+
+Ambient mode — proportional quality enforcement without explicit command invocation.
+
+## Command
+
+### `/ambient`
+
+Classify user intent and apply proportional skill enforcement to any prompt.
+
+```bash
+/ambient add a login form          # BUILD/STANDARD — loads TDD + implementation-patterns
+/ambient fix the auth error        # DEBUG/STANDARD — loads test-patterns + core-patterns
+/ambient where is the config?      # EXPLORE/QUICK — responds normally, zero overhead
+/ambient refactor the auth system  # BUILD/ESCALATE — suggests /implement
+```
+
+## Always-On Mode
+
+Enable ambient classification on every prompt without typing `/ambient`:
+
+```bash
+devflow ambient --enable    # Register UserPromptSubmit hook
+devflow ambient --disable   # Remove hook
+devflow ambient --status    # Check if enabled
+```
+
+When enabled, a `UserPromptSubmit` hook injects a classification preamble before every prompt. Slash commands (`/implement`, `/code-review`, etc.) and short confirmations ("yes", "ok") are skipped automatically.
+
+## How It Works
+
+1. **Classify intent** — BUILD, DEBUG, REVIEW, PLAN, EXPLORE, or CHAT
+2. **Classify depth** — QUICK (zero overhead), STANDARD (2-3 skills), or ESCALATE (workflow nudge)
+3. **Apply proportionally**:
+   - QUICK: respond normally
+   - STANDARD: load relevant skills, enforce TDD for BUILD
+   - ESCALATE: respond + recommend full workflow command
+
+## Depth Tiers
+
+| Depth | When | Overhead |
+|-------|------|----------|
+| QUICK | Chat, exploration, < 20 words, no file refs | ~0 tokens |
+| STANDARD | BUILD/DEBUG/REVIEW/PLAN, 1-5 file scope | ~500-1000 tokens (skill reads) |
+| ESCALATE | Multi-file, architectural, system-wide scope | ~0 extra tokens (nudge only) |
+
+## Skills
+
+- `ambient-router` — Intent + depth classification, skill selection matrix
@@ -0,0 +1,110 @@
+---
+description: Ambient mode — classify intent and apply proportional quality enforcement to any prompt
+---
+
+# Ambient Command
+
+Classify user intent and apply proportional quality enforcement. No agents spawned — enhances the main session only.
+
+## Usage
+
+```
+/ambient <prompt>           Classify and respond with skill enforcement
+/ambient                    Show usage
+```
+
+## Phases
+
+### Phase 1: Load Router
+
+Read the `ambient-router` skill:
+- `~/.claude/skills/ambient-router/SKILL.md`
+
+### Phase 2: Classify
+
+Apply the ambient-router classification to `$ARGUMENTS`:
+
+1. **Intent:** BUILD | DEBUG | REVIEW | PLAN | EXPLORE | CHAT
+2. **Depth:** QUICK | STANDARD | ESCALATE
+
+If no arguments provided, output:
+
+```
+## Ambient Mode
+
+Classify intent and apply proportional quality enforcement.
+
+Usage: /ambient <your prompt>
+
+Examples:
+  /ambient add a login form          → BUILD/STANDARD (loads TDD + implementation-patterns)
+  /ambient fix the auth error        → DEBUG/STANDARD (loads test-patterns + core-patterns)
+  /ambient where is the config?      → EXPLORE/QUICK (responds normally)
+  /ambient refactor the auth system  → BUILD/ESCALATE (suggests /implement)
+
+Always-on: devflow ambient --enable
+```
+
+Then stop.
+
+### Phase 3: State Classification
+
+- **QUICK:** Skip this phase entirely. Respond directly in Phase 4.
+- **STANDARD:** Output one line: `Ambient: {INTENT}/{DEPTH}. Loading: {skill1}, {skill2}.`
+- **ESCALATE:** Skip — recommendation happens in Phase 4.
+
+### Phase 4: Apply
+
+**QUICK:**
+Respond to the user's prompt normally. Zero skill loading. Zero overhead.
+
+**STANDARD:**
+Read the selected skills based on the ambient-router's skill selection matrix:
+
+| Intent | Primary Skills | Secondary (conditional) |
+|--------|---------------|------------------------|
+| BUILD | test-driven-development, implementation-patterns | typescript (.ts), react (.tsx), frontend-design (CSS/UI), input-validation (forms/API), security-patterns (auth/crypto) |
+| DEBUG | test-patterns, core-patterns | git-safety (if git ops) |
+| REVIEW | self-review, core-patterns | test-patterns |
+| PLAN | implementation-patterns | core-patterns |
+
+Read up to 3 skills from `~/.claude/skills/{name}/SKILL.md`. Apply their patterns and constraints when responding to the user's prompt.
+
+For BUILD intent: enforce RED-GREEN-REFACTOR from test-driven-development. Write failing tests before production code.
+
+**ESCALATE:**
+Respond to the user's prompt with your best effort, then append:
+
+> This task spans multiple files/systems. Consider `/implement` for full lifecycle management (exploration → planning → implementation → review).
+
+## Architecture
+
+```
+/ambient <prompt> (main session, no agents)
+│
+├─ Phase 1: Load ambient-router skill
+├─ Phase 2: Classify intent + depth
+├─ Phase 3: State classification (STANDARD only)
+└─ Phase 4: Apply
+   ├─ QUICK → respond directly
+   ├─ STANDARD → load 2-3 skills, apply patterns, respond
+   └─ ESCALATE → respond + workflow nudge
+```
+
+## Edge Cases
+
+| Case | Handling |
+|------|----------|
+| No arguments | Show usage and stop |
+| Single word ("help") | Classify — likely CHAT/QUICK |
+| Prompt references `/implement` etc. | Classify as normal — user chose /ambient intentionally |
+| Mixed intent ("fix and add test") | Use higher-overhead intent (BUILD > DEBUG) |
+| User says "no enforcement" | Respect immediately — treat as QUICK |
+
+## Principles
+
+1. **No agents** — Ambient enhances the main session, never spawns subagents
+2. **Proportional** — QUICK gets zero overhead, STANDARD gets 2-3 skills, ESCALATE gets a nudge
+3. **Transparent** — State classification for STANDARD/ESCALATE, silent for QUICK
+4. **Respectful** — Never over-classify; when in doubt, go one tier lower
+5. **TDD for BUILD** — STANDARD depth BUILD tasks enforce test-first workflow
@@ -21,6 +21,7 @@
     "github-patterns",
     "input-validation",
     "react",
+    "test-driven-development",
     "test-patterns",
     "typescript"
   ]

@@ -0,0 +1,48 @@
+#!/bin/bash
+
+# Ambient Mode: UserPromptSubmit Hook
+# Injects a classification preamble before every user prompt so Claude applies
+# proportional quality enforcement via the ambient-router skill.
+# Zero file I/O beyond stdin — static injection only.
+
+set -euo pipefail
+
+# jq is required to parse hook input JSON — silently no-op if missing
+if ! command -v jq &>/dev/null; then exit 0; fi
+
+INPUT=$(cat)
+
+CWD=$(echo "$INPUT" | jq -r '.cwd // ""' 2>/dev/null)
+if [ -z "$CWD" ]; then
+  exit 0
+fi
+
+PROMPT=$(echo "$INPUT" | jq -r '.prompt // ""' 2>/dev/null)
+
+# Skip slash commands — they have their own orchestration
+if [[ "$PROMPT" == /* ]]; then
+  exit 0
+fi
+
+# Skip short confirmations (< 3 words)
+WORD_COUNT=$(echo "$PROMPT" | wc -w | tr -d ' ')
+if [ "$WORD_COUNT" -lt 3 ]; then
+  exit 0
+fi
+
+# Inject classification preamble
+PREAMBLE="AMBIENT MODE ACTIVE: Before responding, silently classify this prompt:
+Intent: BUILD | DEBUG | REVIEW | PLAN | EXPLORE | CHAT
+Depth: QUICK (no overhead) | STANDARD (load skills) | ESCALATE (suggest /command)
+
+If STANDARD+: Read the ambient-router skill for classification details and skill selection matrix. For BUILD tasks, also load test-driven-development skill and enforce RED-GREEN-REFACTOR.
+
+If QUICK: Respond normally without stating classification.
+Only state classification aloud for STANDARD/ESCALATE."
+
+jq -n --arg ctx "$PREAMBLE" '{
+  "hookSpecificOutput": {
+    "hookEventName": "UserPromptSubmit",
+    "additionalContext": $ctx
+  }
+}'
@@ -102,11 +102,44 @@ fi
 
 # Build instruction
 if [ -n "$EXISTING_MEMORY" ]; then
-  INSTRUCTION="Update the file $MEMORY_FILE with working memory from this session. The file already has content — possibly from a concurrent session that just wrote it moments ago. Merge this session's context with the existing content to produce a single unified working memory snapshot. Both this session and the existing content represent fresh, concurrent work — integrate both fully. Working memory captures what's active now, not a changelog. Deduplicate overlapping information. Keep under 100 lines total. Use the same structure: ## Now, ## Decisions, ## Modified Files, ## Context, ## Session Log.
+  PATTERNS_INSTRUCTION=""
+PATTERNS_FILE="$CWD/.docs/patterns.md"
+EXISTING_PATTERNS=""
+if [ -f "$PATTERNS_FILE" ]; then
+  EXISTING_PATTERNS=$(cat "$PATTERNS_FILE")
+  PATTERNS_INSTRUCTION="
+
+Also update $PATTERNS_FILE by APPENDING any new recurring patterns discovered during this session. Do NOT overwrite existing entries — only add new ones. Skip if no new patterns were observed. Format each entry as: - **Pattern name**: Brief description (discovered: YYYY-MM-DD)
+
+Existing patterns:
+$EXISTING_PATTERNS"
+else
+  PATTERNS_INSTRUCTION="
+
+If recurring patterns were observed during this session (coding conventions, architectural decisions, team preferences, tooling quirks), create $PATTERNS_FILE with entries formatted as: - **Pattern name**: Brief description (discovered: YYYY-MM-DD). Only create this file if genuine patterns were observed — do not fabricate entries."
+fi
+
+INSTRUCTION="Update the file $MEMORY_FILE with working memory from this session. The file already has content — possibly from a concurrent session that just wrote it moments ago. Merge this session's context with the existing content to produce a single unified working memory snapshot. Both this session and the existing content represent fresh, concurrent work — integrate both fully. Working memory captures what's active now, not a changelog. Deduplicate overlapping information. Keep under 100 lines total. Use the same structure: ## Now, ## Decisions, ## Modified Files, ## Context, ## Session Log.${PATTERNS_INSTRUCTION}
 
 Existing content:
 $EXISTING_MEMORY"
 else
+  PATTERNS_INSTRUCTION=""
+  PATTERNS_FILE="$CWD/.docs/patterns.md"
+  if [ -f "$PATTERNS_FILE" ]; then
+    EXISTING_PATTERNS=$(cat "$PATTERNS_FILE")
+    PATTERNS_INSTRUCTION="
+
+Also update $PATTERNS_FILE by APPENDING any new recurring patterns discovered during this session. Do NOT overwrite existing entries — only add new ones. Skip if no new patterns were observed. Format each entry as: - **Pattern name**: Brief description (discovered: YYYY-MM-DD)
+
+Existing patterns:
+$EXISTING_PATTERNS"
+  else
+    PATTERNS_INSTRUCTION="
+
+If recurring patterns were observed during this session (coding conventions, architectural decisions, team preferences, tooling quirks), create $PATTERNS_FILE with entries formatted as: - **Pattern name**: Brief description (discovered: YYYY-MM-DD). Only create this file if genuine patterns were observed — do not fabricate entries."
+  fi
+
   INSTRUCTION="Create the file $MEMORY_FILE with working memory from this session. Keep under 100 lines. Use this structure:
 
 # Working Memory
@@ -129,7 +162,7 @@ else
 <!-- Chronological summary of work done today (2-5 bullets) -->
 
 ### This Week
-<!-- Broader multi-day context if relevant -->"
+<!-- Broader multi-day context if relevant -->${PATTERNS_INSTRUCTION}"
 fi
 
 # Resume session headlessly to perform the update