liteagents 2.4.6 → 2.5.0

package/packages/opencode/command/friction.md

@@ -0,0 +1,139 @@
+---
+name: friction
+description: Analyze session behavior patterns
+usage: /friction <sessions-path>
+argument-hint: <sessions-path>
+---
+
+Analyze Claude Code session logs for friction signals, behavioral patterns, and failure antigens.
+
+**Guardrails**
+- Favor straightforward, minimal implementations first and add complexity only when requested or clearly required.
+- Keep changes tightly scoped to the requested outcome.
+
+**Argument: sessions-path (required)**
+
+Path to the OpenCode sessions directory containing JSONL files.
+- Example: `/friction ~/.config/opencode/projects/-home-hamr-PycharmProjects-aurora/`
+- If no argument provided, ask the user for the path. Do NOT guess.
+
+**What it does**
+
+Runs friction analysis on session JSONL files and writes results to `.opencode/friction/` in the current project.
+
+**Steps**
+
+1. **Validate path**
+   - The argument `$ARGUMENTS` is the sessions path
+   - Verify the path exists and contains `.jsonl` files
+   - If path doesn't exist or has no JSONL files, tell the user and stop
+   - Count total session files found
+
+2. **Run friction.js** (bundled at `commands/friction/friction.js`)
+   - Look for friction.js in order:
+     1. `~/.config/opencode/command/friction/friction.js` (installed)
+     2. `packages/opencode/command/friction/friction.js` (package development)
+   - If found, run: `node <path-to-friction.js> "$ARGUMENTS"`
+   - If it does NOT exist anywhere, fall back to running the analysis manually (step 3)
+   - If friction.js succeeds, skip to step 7
+
+3. **Manual analysis fallback** (only if friction.js not available)
+   - For each `.jsonl` session file in the path:
+     - Read the file
+     - Extract signals using these patterns:
+
+   | Signal | Weight | How to detect |
+   |---|---|---|
+   | `user_intervention` | 10 | User message contains `/stash` |
+   | `session_abandoned` | 10 | Last 3 turns have friction > 15 and no `exit_success` |
+   | `false_success` | 8 | LLM text contains "done"/"complete"/"fixed" AND next tool result has error |
+   | `no_resolution` | 8 | Session has `exit_error` signals but no `exit_success` after them |
+   | `tool_loop` | 6 | Same tool called 3+ times with identical arguments |
+   | `rapid_exit` | 6 | <3 turns AND ends with error |
+   | `interrupt_cascade` | 5 | Multiple `request_interrupted` within 60 seconds |
+   | `user_curse` | 5 | User message matches profanity patterns |
+   | `request_interrupted` | 2.5 | Turn has `is_interrupted: true` or ESC/Ctrl+C signal |
+   | `exit_error` | 1 | Tool result has non-zero exit code |
+   | `repeated_question` | 1 | User asks same question twice (fuzzy match) |
+   | `long_silence` | 0.5 | >10 minute gap between turns |
+   | `user_negation` | 0.5 | User message starts with "no", "wrong", "didn't work" |
+   | `compaction` | 0.5 | System message indicates context compaction |
+
+4. **Score each session**
+   - Accumulate weighted signal scores (no subtraction, only accumulation)
+   - Track peak friction score
+   - Classify session quality:
+     - **BAD**: has `user_intervention` or `session_abandoned`
+     - **FRICTION**: has `user_curse` or `false_success`
+     - **ROUGH**: peak friction >= 15, no intervention
+     - **OK**: low friction, completed normally
+     - **ONE-SHOT**: single turn, not interactive (filter out)
+
+5. **Aggregate stats**
+   - Count sessions by quality
+   - Calculate BAD rate
+   - Identify worst and best sessions
+   - Daily trend if sessions span multiple days
+
+6. **Extract antigen candidates from BAD sessions**
+   - For each BAD session, extract:
+     - Anchor signal (what triggered BAD classification)
+     - Tool sequence around the failure
+     - Error messages
+     - Pattern description
+   - Write as antigen candidates
+
+7. **Write output to `.opencode/friction/`**
+   - Create `.opencode/friction/` directory if it doesn't exist
+   - Write these files:
+     - `friction_analysis.json` — per-session breakdown (quality, peak, signals)
+     - `friction_summary.json` — aggregate stats, verdict, daily trend
+     - `friction_raw.jsonl` — all raw signals with timestamps
+     - `antigen_candidates.json` — raw extracted failure patterns
+     - `antigen_clusters.json` — clustered patterns (primary artifact for /remember)
+     - `antigen_review.md` — human-readable clustered review
+
+8. **Report to user**
+   - Sessions analyzed count
+   - BAD / FRICTION / ROUGH / OK counts
+   - BAD rate percentage
+   - Worst session ID and peak friction
+   - Path to `antigen_review.md` for review
+   - Remind: run `/remember` to consolidate into project memory
+
+**Output format for antigen_review.md**
+
+```markdown
+# Friction Antigen Clusters
+
+Generated: [date]
+BAD sessions: [count] | Raw candidates: [count] | Clusters: [count]
+
+## Cluster Summary
+
+| # | Signal | Tool Pattern | Count | Sessions | Score | Median Peak |
+|---|--------|-------------|-------|----------|-------|-------------|
+| 1 | false_success | Bash,Bash | 35 | 23 | 280 | 73 |
+| 2 | user_intervention | (none) | 34 | 24 | 340 | 65 |
+
+## Cluster 1: false_success | Bash,Bash
+
+**Occurrences:** 35 across 23 sessions | **Score:** 280
+
+### User Context (what the user said)
+> [actual user quote from session]
+
+### Errors
+[error messages if any]
+
+### Files involved
+- [file paths]
+```
+
+**File locations**
+- Design doc: `docs/02-features/memory/LONG_TERM_MEMORY.md`
+- Input: JSONL session files from `$ARGUMENTS` path
+- Output: `.opencode/friction/` (project-local)
+- JS script: `commands/friction/friction.js` (bundled alongside this command)
+- Python equivalent: `scripts/friction.py` (aurora users, same logic)
+- Manual fallback: built into this command if friction.js unavailable

package/packages/opencode/command/remember.md

@@ -0,0 +1,110 @@
+---
+name: remember
+description: Consolidate stashes + friction into project memory
+usage: /remember
+---
+
+Consolidate session stashes and friction analysis into a single project-local MEMORY.md, then inject into AGENTS.md.
+
+**Guardrails**
+- Favor straightforward, minimal implementations first and add complexity only when requested or clearly required.
+- Keep changes tightly scoped to the requested outcome.
+
+**What it does**
+
+Reads all raw material (`.opencode/stash/*.md` + `.opencode/friction/antigen_clusters.json`), extracts durable facts, episodes, and behavioral preferences into a single `.opencode/memory/MEMORY.md`, then injects a managed memory section into `AGENTS.md`.
+
+**Steps**
+
+1. **Gather sources**
+   - Read all `.opencode/stash/*.md` files in the current project
+   - Check for friction output at `.opencode/friction/antigen_clusters.json` (preferred) or `.opencode/friction/antigen_review.md` (fallback)
+   - Read existing `.opencode/memory/MEMORY.md` if it exists — create dir if missing
+   - Read processed manifest at `.opencode/memory/.processed` — skip already-processed stashes
+   - If no unprocessed stashes, report "nothing to consolidate" and stop
+
+2. **Extract from unprocessed stashes** (use Task tool with sonnet model for each)
+   - For each unprocessed stash, call sonnet to extract:
+     - **FACTS** (atomic, one-line): stable preferences, decisions, corrections, explicit "remember this"
+     - **EPISODE** (3-5 bullet narrative): what was the goal, what was tried, outcome, lesson
+     - **SKIP**: code details, file paths, errors, mechanical steps, LLM responses
+   - Collect all new facts and episodes
+
+3. **Merge into MEMORY.md**
+   - Read existing `.opencode/memory/MEMORY.md` and parse its three sections (## Facts, ## Episodes, ## Preferences)
+   - **Facts section**: call sonnet with existing facts + newly extracted facts
+     - Rules: new updates replace old, contradictions keep new version, duplicates dropped
+     - Keep facts atomic, one line each
+   - **Episodes section**: append new episode entries (append-only, timestamped, no dedup)
+   - **Preferences section**: only update if friction output exists (step 4)
+   - Write merged result to `.opencode/memory/MEMORY.md` in this format:
+
+   ```markdown
+   # Project Memory
+   > Auto-generated by /remember. Do not edit manually.
+
+   ## Facts
+   - [atomic fact 1]
+   - [atomic fact 2]
+
+   ## Episodes
+   ### YYYY-MM-DD - [title]
+   - [bullet narrative]
+
+   ### YYYY-MM-DD - [title]
+   - [bullet narrative]
+
+   ## Preferences
+   ### High Confidence
+   - [pattern] (evidence: [count] observations)
+
+   ### Medium Confidence
+   - [pattern] (evidence: [count] observations)
+
+   ### Low Confidence
+   - [pattern] (evidence: [count] observations)
+   ```
+
+4. **Distill friction into preferences** (only if friction output exists)
+   - Read `.opencode/friction/antigen_clusters.json` — this contains clustered failure patterns with counts, user_context quotes, and tool sequences
+   - For the **top 10 clusters** (by score), extract the `contexts` field — these are the actual user messages that show behavioral patterns
+   - Call sonnet with:
+     - The top 10 clusters (signal, tool_pattern, count, sessions, contexts, errors)
+     - Existing Preferences section from MEMORY.md
+   - Extract BEHAVIORAL preferences (patterns demonstrated, not stated) — the user_context quotes are the primary evidence
+   - Confidence tiers:
+     - **High Confidence**: 5+ observations — available via @MEMORY.md
+     - **Medium Confidence**: 3+ observations — observing, not loaded
+     - **Low Confidence**: <3 observations — needs more data
+   - Promote/demote based on new evidence
+   - Update the Preferences section in MEMORY.md
+
+5. **Inject memory reference into AGENTS.md**
+   - Compose the section between `<!-- MEMORY:START -->` and `<!-- MEMORY:END -->` markers:
+     ```
+     <!-- MEMORY:START -->
+     @MEMORY.md
+     <!-- MEMORY:END -->
+     ```
+   - The `@MEMORY.md` reference points to `.opencode/memory/MEMORY.md` — Claude loads the full file directly, so no inline duplication is needed
+   - If AGENTS.md already has MEMORY markers, replace the section between them
+   - If AGENTS.md has no MEMORY markers, append the section at the end
+   - If no AGENTS.md exists, create one with just the memory section
+
+6. **Update processed manifest**
+   - Append paths of newly processed stashes to `.opencode/memory/.processed`
+
+7. **Report to user**
+   - Number of stashes processed
+   - Facts count (total, new)
+   - Episodes count (total, new)
+   - Preferences count by confidence tier
+   - Confirm MEMORY.md and AGENTS.md updated
+
+**File locations (all project-local)**
+- Memory file: `.opencode/memory/MEMORY.md` (single source of truth, referenced as @MEMORY.md)
+- Stash files: `.opencode/stash/*.md`
+- Friction output: `.opencode/friction/antigen_clusters.json` (clustered patterns with user contexts)
+- Friction fallback: `.opencode/friction/antigen_review.md` (human-readable clusters)
+- Processed manifest: `.opencode/memory/.processed`
+- Output: `AGENTS.md` (managed MEMORY section with @MEMORY.md reference)

package/packages/opencode/opencode.jsonc

@@ -185,6 +185,14 @@
       "template": "{file:./command/ship.md}",
       "description": "Pre-deployment verification checklist"
     },
+    "friction": {
+      "template": "{file:./command/friction.md}",
+      "description": "Analyze session logs for failure patterns and behavioral signals"
+    },
+    "remember": {
+      "template": "{file:./command/remember.md}",
+      "description": "Consolidate stashes + friction into project memory"
+    },
     "stash": {
       "template": "{file:./command/stash.md}",
       "description": "Save session context for compaction recovery or handoffs"

package/packages/subagentic-manual.md

@@ -31,10 +31,10 @@ Install for your platform:
 
 | Platform | Installation | What's Included |
 |----------|--------------|-----------------|
-| **Claude Code** | `cp -r claude/* ~/.claude/` | 11 subagents + 10 skills + 10 commands |
-| **Droid** | `cp -r droid/* ~/.factory/` | 20 commands (subagent references) |
-| **Ampcode** | `cp -r ampcode/* ~/.config/amp/` | 11 subagents + 10 skills + 10 commands |
-| **OpenCode** | `cp -r opencode/* ~/.config/opencode/` | 20 commands (subagent references) |
+| **Claude Code** | `cp -r claude/* ~/.claude/` | 11 subagents + 10 skills + 12 commands |
+| **Droid** | `cp -r droid/* ~/.factory/` | 22 commands (subagent references) |
+| **Ampcode** | `cp -r ampcode/* ~/.config/amp/` | 11 subagents + 10 skills + 12 commands |
+| **OpenCode** | `cp -r opencode/* ~/.config/opencode/` | 22 commands (subagent references) |
 
 **Key Difference**:
 - **Claude Code / Ampcode** implement full subagent system with orchestrator
@@ -54,8 +54,8 @@ Install for your platform:
 - test-driven-development, testing-anti-patterns, verification-before-completion (auto-trigger)
 - brainstorming, code-review, systematic-debugging, docs-builder, etc.
 
-**10 Commands** - Simple workflow helpers
-- debug, explain, git-commit, optimize, refactor, review, security, ship, stash, test-generate
+**12 Commands** - Simple workflow helpers
+- debug, explain, friction, git-commit, optimize, refactor, remember, review, security, ship, stash, test-generate
 
 **Orchestration System**
 - Automatic intent matching to 9 workflow patterns
@@ -64,7 +64,7 @@ Install for your platform:
 
 ### Droid/OpenCode (Commands Only)
 
-**20 Commands** - All workflow capabilities in command form
+**22 Commands** - All workflow capabilities in command form
 - Combines skills + commands into unified command set
 - Same functionality, different invocation model (no auto-triggering)
 - Includes reference documentation for subagents
@@ -102,7 +102,7 @@ Install for your platform:
 
 ## Commands Reference
 
-### Claude Code / Ampcode: 20 Total (10 Skills + 10 Commands)
+### Claude Code / Ampcode: 22 Total (10 Skills + 12 Commands)
 
 **Auto-Triggering Skills (3)**
 - `test-driven-development` - Write test first, watch fail, minimal passing code
@@ -118,19 +118,21 @@ Install for your platform:
 - `skill-creator` - Guide for creating new skills
 - `systematic-debugging` - Four-phase debugging framework
 
-**Simple Commands (10)**
+**Simple Commands (12)**
 - `debug` - Systematic investigation techniques
 - `explain` - Explain code for newcomers
+- `friction` - Analyze session logs for failure patterns and behavioral signals
 - `git-commit` - Intelligent commit creation
 - `optimize` - Performance analysis
 - `refactor` - Maintain behavior while improving code
+- `remember` - Consolidate stashes + friction into project memory
 - `review` - Comprehensive code review
 - `security` - Vulnerability scanning
 - `ship` - Pre-deployment checklist
 - `stash` - Save session context for compaction recovery or handoffs
 - `test-generate` - Test suite generation
 
-### Droid/OpenCode: 20 Commands
+### Droid/OpenCode: 22 Commands
 
 Same functionality as skills+commands, but:
 - All invoked as commands (no auto-triggering)
@@ -140,7 +142,23 @@ Same functionality as skills+commands, but:
 **Command Categories**:
 - **Development & Testing (9)**: test-driven-development, testing-anti-patterns, test-generate, code-review, systematic-debugging, root-cause-tracing, debug, condition-based-waiting, verification-before-completion
 - **Code Operations (6)**: refactor, optimize, explain, review, security, ship
-- **Session & Planning (5)**: brainstorming, skill-creator, docs-builder, git-commit, stash
+- **Session & Memory (7)**: brainstorming, skill-creator, docs-builder, git-commit, stash, friction, remember
+
+---
+
+## Hot Memory
+
+Lightweight session memory that learns from your usage patterns across sessions.
+
+```
+/stash → /friction → /remember
+```
+
+1. **`/stash`** - Snapshot current session context to `.claude/stash/`. Use before compaction, handoffs, or ending complex work.
+2. **`/friction`** - Analyze session logs for failure patterns. Scores sessions, clusters failures by signal and tool sequence, outputs actionable antigens to `.claude/friction/`.
+3. **`/remember`** - Consolidate stashes + friction into `.claude/memory/MEMORY.md`. Extracts facts, episodes, and behavioral preferences. References into CLAUDE.md via `@MEMORY.md`.
+
+**Result:** Project-local memory that accumulates across sessions — no external dependencies, no databases, just markdown.
 
 ---
 
@@ -237,7 +255,7 @@ Subagent workflows require manual coordination.
 ├── CLAUDE.md           # Registry + orchestrator workflows
 ├── agents/             # 11 subagent implementations (*.md)
 ├── skills/             # 11 skills (subdirectories with SKILL.md)
-└── commands/           # 10 commands (*.md)
+└── commands/           # 12 commands (*.md)
 ```
 
 **Features**:
@@ -252,7 +270,7 @@ Subagent workflows require manual coordination.
 ├── AGENT.md            # Reference doc (subagents + commands)
 ├── agents/             # 11 subagent implementations (*.md)
 ├── skills/             # 11 skills (subdirectories with SKILL.md)
-└── commands/           # 10 commands (*.md)
+└── commands/           # 12 commands (*.md)
 ```
 
 **Features**:
@@ -264,7 +282,7 @@ Subagent workflows require manual coordination.
 ```
 ~/.factory/
 ├── AGENTS.md           # Reference doc (subagents + commands)
-└── commands/           # 20 commands (*.md)
+└── commands/           # 22 commands (*.md)
 ```
 
 **Features**:
@@ -276,7 +294,7 @@ Subagent workflows require manual coordination.
 ```
 ~/.config/opencode/
 ├── AGENTS.md           # Reference doc (subagents + commands)
-└── command/            # 20 commands (*.md)
+└── command/            # 22 commands (*.md)
 ```
 
 **Features**:

package/packages/ampcode/README.md

@@ -1,17 +0,0 @@
-# Ampcode Packages
-
-Ampcode AI codegen tool packages for amplified development workflows.
-
-## Structure
-- `lite/` - Basic amplified codegen setup
-- `standard/` - Enhanced automation with amplified workflows
-- `pro/` - Maximum amplification with advanced AI codegen capabilities
-
-## Optimization Focus
-- Enhanced automation capabilities
-- Amplified development workflows
-- Accelerated AI codegen
-- Advanced development patterns
-
-## Status: Placeholder
-Content to be developed based on Claude packages with amplification-specific optimizations.

package/packages/claude/README.md

@@ -1,23 +0,0 @@
-# Claude Code Packages
-
-Claude Code AI development assistant packages optimized for conversational AI interactions.
-
-## Structure
-- `lite/` - Minimal setup with 3 core agents
-- `standard/` - Full agent set with 8 core skills  
-- `pro/` - Complete toolkit with 14 advanced skills
-
-## Content Source
-Existing content available in `../claude_temp_backup/` ready to be organized into variants.
-
-## Optimization Focus
-- Conversational AI patterns
-- Markdown formatting
-- Claude-specific response handling
-- Plugin integration
-
-## Next Steps
-1. Copy 3 core agents to `lite/`
-2. Copy all agents + 8 skills to `standard/`
-3. Copy all agents + 14 skills to `pro/`
-4. Customize for Claude-specific optimizations

package/packages/droid/README.md

@@ -1,17 +0,0 @@
-# Droid Packages
-
-Droid AI codegen tool packages specialized for mobile development.
-
-## Structure
-- `lite/` - Basic mobile codegen setup
-- `standard/` - Mobile-focused agents with core development skills
-- `pro/` - Complete mobile AI codegen toolkit with platform integration
-
-## Optimization Focus
-- Mobile development patterns
-- Android/iOS codegen
-- Platform-specific workflows
-- Mobile-first development
-
-## Status: Placeholder
-Content to be developed based on Claude packages with mobile-specific optimizations.

package/packages/opencode/README.md

@@ -1,17 +0,0 @@
-# Opencode Packages
-
-Opencode AI codegen tool packages optimized for CLI-based development workflows.
-
-## Structure
-- `lite/` - Minimal setup for basic CLI codegen
-- `standard/` - Full CLI-optimized agent set with core skills
-- `pro/` - Complete CLI codegen toolkit with advanced automation
-
-## Optimization Focus
-- CLI-based interactions
-- Terminal workflow integration
-- Command-line development patterns
-- Accelerated CLI codegen
-
-## Status: Placeholder
-Content to be developed based on Claude packages with CLI-specific optimizations.