squads-cli 0.2.0 → 0.4.0

package/docker/init-db.sql

@@ -0,0 +1,399 @@
+-- Squads Local Database Initialization
+-- Creates additional schemas/tables beyond what Langfuse creates
+
+-- Schema for squads-specific data (separate from Langfuse)
+CREATE SCHEMA IF NOT EXISTS squads;
+
+-- GitHub metrics table - tracks git activity over time
+CREATE TABLE IF NOT EXISTS squads.github_metrics (
+    id SERIAL PRIMARY KEY,
+    org VARCHAR(255) NOT NULL,
+    repo VARCHAR(255) NOT NULL,
+    metric_date DATE NOT NULL,
+
+    -- Activity counts
+    commits INTEGER DEFAULT 0,
+    prs_opened INTEGER DEFAULT 0,
+    prs_merged INTEGER DEFAULT 0,
+    prs_closed INTEGER DEFAULT 0,
+    issues_opened INTEGER DEFAULT 0,
+    issues_closed INTEGER DEFAULT 0,
+    reviews INTEGER DEFAULT 0,
+
+    -- Code metrics
+    additions INTEGER DEFAULT 0,
+    deletions INTEGER DEFAULT 0,
+
+    -- Quality signals
+    avg_pr_cycle_hours NUMERIC(10,2),
+    review_pass_rate NUMERIC(5,2),
+
+    -- Metadata
+    created_at TIMESTAMP WITH TIME ZONE DEFAULT NOW(),
+
+    UNIQUE(org, repo, metric_date)
+);
+
+-- Agent execution metrics - links to Langfuse traces
+CREATE TABLE IF NOT EXISTS squads.agent_executions (
+    id SERIAL PRIMARY KEY,
+    squad VARCHAR(100) NOT NULL,
+    agent VARCHAR(100) NOT NULL,
+    execution_id VARCHAR(255), -- Links to Langfuse trace_id
+
+    -- Timing
+    started_at TIMESTAMP WITH TIME ZONE NOT NULL,
+    completed_at TIMESTAMP WITH TIME ZONE,
+    duration_ms INTEGER,
+
+    -- Results
+    status VARCHAR(50) DEFAULT 'running', -- running, completed, failed
+    output_type VARCHAR(50), -- commit, pr, issue, file, etc
+    output_ref VARCHAR(255), -- PR URL, commit SHA, etc
+
+    -- Cost tracking (backup to Langfuse)
+    input_tokens INTEGER DEFAULT 0,
+    output_tokens INTEGER DEFAULT 0,
+    cost_usd NUMERIC(10,6) DEFAULT 0,
+
+    -- Metadata
+    created_at TIMESTAMP WITH TIME ZONE DEFAULT NOW()
+);
+
+-- Baseline snapshots - for before/after comparison
+CREATE TABLE IF NOT EXISTS squads.baselines (
+    id SERIAL PRIMARY KEY,
+    org VARCHAR(255) NOT NULL,
+    name VARCHAR(100) NOT NULL,
+    captured_at TIMESTAMP WITH TIME ZONE DEFAULT NOW(),
+
+    -- 30-day metrics at capture time
+    commits_30d INTEGER DEFAULT 0,
+    prs_30d INTEGER DEFAULT 0,
+    issues_30d INTEGER DEFAULT 0,
+    avg_pr_cycle_hours NUMERIC(10,2),
+
+    -- Metadata
+    notes TEXT,
+
+    UNIQUE(org, name)
+);
+
+-- Create indexes for common queries
+CREATE INDEX IF NOT EXISTS idx_github_metrics_date ON squads.github_metrics(metric_date DESC);
+CREATE INDEX IF NOT EXISTS idx_github_metrics_repo ON squads.github_metrics(org, repo);
+CREATE INDEX IF NOT EXISTS idx_agent_executions_squad ON squads.agent_executions(squad, agent);
+CREATE INDEX IF NOT EXISTS idx_agent_executions_status ON squads.agent_executions(status);
+
+-- Dashboard snapshots - full dashboard state over time
+CREATE TABLE IF NOT EXISTS squads.dashboard_snapshots (
+    id SERIAL PRIMARY KEY,
+    captured_at TIMESTAMP WITH TIME ZONE DEFAULT NOW(),
+
+    -- Top-level metrics
+    total_squads INTEGER DEFAULT 0,
+    total_commits INTEGER DEFAULT 0,
+    total_prs_merged INTEGER DEFAULT 0,
+    total_issues_closed INTEGER DEFAULT 0,
+    total_issues_open INTEGER DEFAULT 0,
+    goal_progress_pct INTEGER DEFAULT 0,
+
+    -- Cost metrics (from Langfuse)
+    cost_usd NUMERIC(10,4) DEFAULT 0,
+    daily_budget_usd NUMERIC(10,2) DEFAULT 50,
+    input_tokens INTEGER DEFAULT 0,
+    output_tokens INTEGER DEFAULT 0,
+
+    -- Git activity
+    commits_30d INTEGER DEFAULT 0,
+    avg_commits_per_day NUMERIC(5,1) DEFAULT 0,
+    active_days INTEGER DEFAULT 0,
+    peak_commits INTEGER DEFAULT 0,
+    peak_date DATE,
+
+    -- Squad breakdown (JSONB for flexibility)
+    squads_data JSONB DEFAULT '[]'::jsonb,
+    -- Format: [{ name, commits, prs, issues_closed, issues_open, goals_active, goals_total, progress }]
+
+    -- Authors breakdown
+    authors_data JSONB DEFAULT '[]'::jsonb,
+    -- Format: [{ name, commits }]
+
+    -- Repos breakdown
+    repos_data JSONB DEFAULT '[]'::jsonb
+    -- Format: [{ name, commits }]
+);
+
+-- Index for time-series queries
+CREATE INDEX IF NOT EXISTS idx_dashboard_snapshots_date ON squads.dashboard_snapshots(captured_at DESC);
+
+-- =============================================================================
+-- Telemetry Tables - Primary data store (Langfuse is optional forwarding)
+-- =============================================================================
+
+-- LLM generations - every API call to Claude/OpenAI
+CREATE TABLE IF NOT EXISTS squads.llm_generations (
+    id SERIAL PRIMARY KEY,
+    session_id VARCHAR(255) NOT NULL,
+    trace_id VARCHAR(255),
+
+    -- Context
+    squad VARCHAR(100) DEFAULT 'hq',
+    agent VARCHAR(100) DEFAULT 'coo',
+    user_id VARCHAR(255),
+
+    -- Model info
+    model VARCHAR(100) NOT NULL,
+
+    -- Token counts
+    input_tokens INTEGER DEFAULT 0,
+    output_tokens INTEGER DEFAULT 0,
+    cache_read_tokens INTEGER DEFAULT 0,
+    cache_creation_tokens INTEGER DEFAULT 0,
+
+    -- Cost
+    cost_usd NUMERIC(10,6) DEFAULT 0,
+
+    -- Timing
+    created_at TIMESTAMP WITH TIME ZONE DEFAULT NOW(),
+    duration_ms INTEGER,
+
+    -- Metadata (prompt summary, response summary, etc)
+    metadata JSONB DEFAULT '{}'::jsonb
+);
+
+-- Tool executions - every tool call
+CREATE TABLE IF NOT EXISTS squads.tool_executions (
+    id SERIAL PRIMARY KEY,
+    session_id VARCHAR(255) NOT NULL,
+    trace_id VARCHAR(255),
+
+    -- Context
+    squad VARCHAR(100) DEFAULT 'hq',
+    agent VARCHAR(100) DEFAULT 'coo',
+
+    -- Tool info
+    tool_name VARCHAR(255) NOT NULL,
+    success BOOLEAN DEFAULT true,
+
+    -- Timing
+    created_at TIMESTAMP WITH TIME ZONE DEFAULT NOW(),
+    duration_ms INTEGER,
+
+    -- Metadata (parameters, result summary)
+    metadata JSONB DEFAULT '{}'::jsonb
+);
+
+-- Sessions - groups related generations/tools
+CREATE TABLE IF NOT EXISTS squads.sessions (
+    id VARCHAR(255) PRIMARY KEY,
+    squad VARCHAR(100) DEFAULT 'hq',
+    agent VARCHAR(100) DEFAULT 'coo',
+    user_id VARCHAR(255),
+
+    -- Aggregated metrics (updated on each event)
+    total_input_tokens INTEGER DEFAULT 0,
+    total_output_tokens INTEGER DEFAULT 0,
+    total_cost_usd NUMERIC(10,6) DEFAULT 0,
+    generation_count INTEGER DEFAULT 0,
+    tool_count INTEGER DEFAULT 0,
+
+    -- Timing
+    started_at TIMESTAMP WITH TIME ZONE DEFAULT NOW(),
+    last_activity_at TIMESTAMP WITH TIME ZONE DEFAULT NOW(),
+
+    -- Status
+    status VARCHAR(50) DEFAULT 'active' -- active, completed, failed
+);
+
+-- Indexes for telemetry queries
+CREATE INDEX IF NOT EXISTS idx_llm_generations_session ON squads.llm_generations(session_id);
+CREATE INDEX IF NOT EXISTS idx_llm_generations_squad ON squads.llm_generations(squad, agent);
+CREATE INDEX IF NOT EXISTS idx_llm_generations_created ON squads.llm_generations(created_at DESC);
+CREATE INDEX IF NOT EXISTS idx_tool_executions_session ON squads.tool_executions(session_id);
+CREATE INDEX IF NOT EXISTS idx_tool_executions_tool ON squads.tool_executions(tool_name);
+CREATE INDEX IF NOT EXISTS idx_sessions_squad ON squads.sessions(squad, agent);
+CREATE INDEX IF NOT EXISTS idx_sessions_status ON squads.sessions(status);
+
+-- =============================================================================
+-- Conversations - Captured from Claude Code sessions (engram hook)
+-- =============================================================================
+
+-- Conversations/memories from Claude Code sessions
+CREATE TABLE IF NOT EXISTS squads.conversations (
+    id SERIAL PRIMARY KEY,
+    session_id VARCHAR(255),
+    user_id VARCHAR(255) DEFAULT 'local',
+
+    -- Content
+    role VARCHAR(50) NOT NULL, -- user, assistant, thinking
+    content TEXT NOT NULL,
+
+    -- Classification
+    message_type VARCHAR(50) DEFAULT 'message', -- message, thinking, decision, learning
+    importance VARCHAR(20) DEFAULT 'normal', -- low, normal, high
+
+    -- Context
+    squad VARCHAR(100),
+    agent VARCHAR(100),
+    working_dir VARCHAR(500),
+
+    -- Timing
+    created_at TIMESTAMP WITH TIME ZONE DEFAULT NOW(),
+
+    -- Metadata (hook info, extracted entities, etc)
+    metadata JSONB DEFAULT '{}'::jsonb
+);
+
+-- Full-text search on content
+CREATE INDEX IF NOT EXISTS idx_conversations_content_search
+    ON squads.conversations USING gin(to_tsvector('english', content));
+CREATE INDEX IF NOT EXISTS idx_conversations_session ON squads.conversations(session_id);
+CREATE INDEX IF NOT EXISTS idx_conversations_user ON squads.conversations(user_id);
+CREATE INDEX IF NOT EXISTS idx_conversations_created ON squads.conversations(created_at DESC);
+CREATE INDEX IF NOT EXISTS idx_conversations_type ON squads.conversations(message_type);
+CREATE INDEX IF NOT EXISTS idx_conversations_importance ON squads.conversations(importance);
+
+-- =============================================================================
+-- CLI Events - Anonymous telemetry from squads-cli
+-- =============================================================================
+
+CREATE TABLE IF NOT EXISTS squads.cli_events (
+    id SERIAL PRIMARY KEY,
+    received_at TIMESTAMP WITH TIME ZONE DEFAULT NOW(),
+
+    -- Anonymous identification
+    anonymous_id VARCHAR(100),
+
+    -- Event data
+    event_name VARCHAR(100) NOT NULL,
+    cli_version VARCHAR(20),
+
+    -- Properties (flexible JSON)
+    properties JSONB DEFAULT '{}'::jsonb
+);
+
+-- Indexes for analytics queries
+CREATE INDEX IF NOT EXISTS idx_cli_events_name ON squads.cli_events(event_name);
+CREATE INDEX IF NOT EXISTS idx_cli_events_received ON squads.cli_events(received_at DESC);
+CREATE INDEX IF NOT EXISTS idx_cli_events_anonymous ON squads.cli_events(anonymous_id);
+
+-- =============================================================================
+-- Task Tracking - Track task completion, retries, and quality
+-- =============================================================================
+
+-- Tasks - track goal/task completion
+CREATE TABLE IF NOT EXISTS squads.tasks (
+    id SERIAL PRIMARY KEY,
+    task_id VARCHAR(255) UNIQUE NOT NULL, -- External ID for deduplication
+    session_id VARCHAR(255),
+
+    -- Context
+    squad VARCHAR(100) NOT NULL,
+    agent VARCHAR(100),
+
+    -- Task info
+    task_type VARCHAR(50) DEFAULT 'goal', -- goal, issue, pr, command
+    description TEXT,
+
+    -- Completion tracking
+    status VARCHAR(50) DEFAULT 'started', -- started, completed, failed, cancelled
+    success BOOLEAN,
+    retry_count INTEGER DEFAULT 0,
+
+    -- Output
+    output_type VARCHAR(50), -- commit, pr, issue, file, none
+    output_ref VARCHAR(500), -- URL or reference
+
+    -- Cost
+    total_tokens INTEGER DEFAULT 0,
+    total_cost_usd NUMERIC(10,6) DEFAULT 0,
+
+    -- Context window
+    peak_context_tokens INTEGER DEFAULT 0, -- Max context size during task
+    context_utilization_pct NUMERIC(5,2), -- Peak % of context window used
+
+    -- Timing
+    started_at TIMESTAMP WITH TIME ZONE DEFAULT NOW(),
+    completed_at TIMESTAMP WITH TIME ZONE,
+    duration_ms INTEGER,
+
+    -- Metadata
+    metadata JSONB DEFAULT '{}'::jsonb
+);
+
+-- User feedback on task quality
+CREATE TABLE IF NOT EXISTS squads.task_feedback (
+    id SERIAL PRIMARY KEY,
+    task_id VARCHAR(255) REFERENCES squads.tasks(task_id),
+
+    -- Rating
+    quality_score INTEGER CHECK (quality_score >= 1 AND quality_score <= 5), -- 1-5 stars
+
+    -- Detailed feedback
+    was_helpful BOOLEAN,
+    required_fixes BOOLEAN DEFAULT false,
+    fix_description TEXT,
+
+    -- Tags
+    tags VARCHAR(50)[], -- ['accurate', 'fast', 'needed-revision', etc]
+
+    -- Timing
+    created_at TIMESTAMP WITH TIME ZONE DEFAULT NOW(),
+
+    -- Free-form notes
+    notes TEXT
+);
+
+-- Aggregated insights per squad/agent (materialized for fast queries)
+CREATE TABLE IF NOT EXISTS squads.agent_insights (
+    id SERIAL PRIMARY KEY,
+    captured_at TIMESTAMP WITH TIME ZONE DEFAULT NOW(),
+    period VARCHAR(20) NOT NULL, -- 'day', 'week', 'month'
+    period_start DATE NOT NULL,
+
+    -- Context
+    squad VARCHAR(100) NOT NULL,
+    agent VARCHAR(100),
+
+    -- Task metrics
+    tasks_started INTEGER DEFAULT 0,
+    tasks_completed INTEGER DEFAULT 0,
+    tasks_failed INTEGER DEFAULT 0,
+    success_rate NUMERIC(5,2), -- percentage
+
+    -- Retry metrics
+    total_retries INTEGER DEFAULT 0,
+    avg_retries_per_task NUMERIC(5,2),
+    tasks_with_retries INTEGER DEFAULT 0,
+
+    -- Quality metrics (from feedback)
+    avg_quality_score NUMERIC(3,2),
+    feedback_count INTEGER DEFAULT 0,
+    helpful_pct NUMERIC(5,2),
+    fix_required_pct NUMERIC(5,2),
+
+    -- Efficiency metrics
+    avg_duration_ms INTEGER,
+    avg_tokens_per_task INTEGER,
+    avg_cost_per_task NUMERIC(10,6),
+    avg_context_utilization NUMERIC(5,2),
+
+    -- Tool usage
+    top_tools JSONB DEFAULT '[]'::jsonb, -- [{name, count, success_rate}]
+    tool_failure_rate NUMERIC(5,2),
+
+    UNIQUE(period, period_start, squad, agent)
+);
+
+-- Indexes for insights queries
+CREATE INDEX IF NOT EXISTS idx_tasks_squad ON squads.tasks(squad, agent);
+CREATE INDEX IF NOT EXISTS idx_tasks_status ON squads.tasks(status);
+CREATE INDEX IF NOT EXISTS idx_tasks_created ON squads.tasks(started_at DESC);
+CREATE INDEX IF NOT EXISTS idx_task_feedback_task ON squads.task_feedback(task_id);
+CREATE INDEX IF NOT EXISTS idx_agent_insights_lookup ON squads.agent_insights(squad, period, period_start DESC);
+
+-- Grant permissions
+GRANT ALL PRIVILEGES ON SCHEMA squads TO squads;
+GRANT ALL PRIVILEGES ON ALL TABLES IN SCHEMA squads TO squads;
+GRANT USAGE, SELECT ON ALL SEQUENCES IN SCHEMA squads TO squads;

package/docker/init-engram-db.sql

@@ -0,0 +1,148 @@
+-- =============================================================================
+-- Engram Database Initialization
+-- =============================================================================
+-- Creates the engram database with pgvector extension for memory storage
+-- This script runs after init-db.sql (squads schema)
+
+-- Create engram database
+SELECT 'CREATE DATABASE engram'
+WHERE NOT EXISTS (SELECT FROM pg_database WHERE datname = 'engram')\gexec
+
+-- Connect to engram database and set up extensions
+\c engram
+
+-- Enable pgvector for vector similarity search
+CREATE EXTENSION IF NOT EXISTS vector;
+
+-- Enable pg_trgm for fuzzy text search
+CREATE EXTENSION IF NOT EXISTS pg_trgm;
+
+-- Create engram schema
+CREATE SCHEMA IF NOT EXISTS engram;
+
+-- =============================================================================
+-- Auth Tables - Token-based authentication for MCP server
+-- =============================================================================
+
+-- Users table
+CREATE TABLE IF NOT EXISTS engram.users (
+    id SERIAL PRIMARY KEY,
+    email VARCHAR(255) UNIQUE NOT NULL,
+    name VARCHAR(255),
+    created_at TIMESTAMP WITH TIME ZONE DEFAULT NOW(),
+    updated_at TIMESTAMP WITH TIME ZONE DEFAULT NOW()
+);
+
+-- API tokens table
+CREATE TABLE IF NOT EXISTS engram.tokens (
+    id SERIAL PRIMARY KEY,
+    user_id INTEGER REFERENCES engram.users(id) ON DELETE CASCADE,
+    token_hash VARCHAR(255) NOT NULL,
+    name VARCHAR(255),
+    last_used_at TIMESTAMP WITH TIME ZONE,
+    expires_at TIMESTAMP WITH TIME ZONE,
+    created_at TIMESTAMP WITH TIME ZONE DEFAULT NOW(),
+    revoked_at TIMESTAMP WITH TIME ZONE
+);
+
+-- Token usage audit log
+CREATE TABLE IF NOT EXISTS engram.token_usage (
+    id SERIAL PRIMARY KEY,
+    token_id INTEGER REFERENCES engram.tokens(id) ON DELETE SET NULL,
+    user_id INTEGER REFERENCES engram.users(id) ON DELETE SET NULL,
+    endpoint VARCHAR(255),
+    ip_address INET,
+    user_agent TEXT,
+    created_at TIMESTAMP WITH TIME ZONE DEFAULT NOW()
+);
+
+-- =============================================================================
+-- Memory Tables - Core memory storage
+-- =============================================================================
+
+-- Note: mem0 creates its own tables for memory storage
+-- These are additional tables for Engram-specific features
+
+-- Memory metadata (extends mem0 memories)
+CREATE TABLE IF NOT EXISTS engram.memory_metadata (
+    id SERIAL PRIMARY KEY,
+    memory_id VARCHAR(255) UNIQUE NOT NULL,  -- Links to mem0 memory
+    user_id INTEGER REFERENCES engram.users(id),
+    project_id VARCHAR(255),
+    importance FLOAT DEFAULT 0.5,
+    access_count INTEGER DEFAULT 0,
+    last_accessed_at TIMESTAMP WITH TIME ZONE,
+    created_at TIMESTAMP WITH TIME ZONE DEFAULT NOW(),
+    metadata JSONB DEFAULT '{}'::jsonb
+);
+
+-- Memory chunks (for large text chunking)
+CREATE TABLE IF NOT EXISTS engram.memory_chunks (
+    id SERIAL PRIMARY KEY,
+    parent_memory_id VARCHAR(255) NOT NULL,
+    chunk_index INTEGER NOT NULL,
+    total_chunks INTEGER NOT NULL,
+    chunk_size INTEGER NOT NULL,
+    has_overlap BOOLEAN DEFAULT false,
+    created_at TIMESTAMP WITH TIME ZONE DEFAULT NOW(),
+    UNIQUE(parent_memory_id, chunk_index)
+);
+
+-- =============================================================================
+-- Graph Sync Tables - Tracks Neo4j synchronization
+-- =============================================================================
+
+CREATE TABLE IF NOT EXISTS engram.graph_sync_log (
+    id SERIAL PRIMARY KEY,
+    memory_id VARCHAR(255) NOT NULL,
+    sync_type VARCHAR(50) NOT NULL,  -- 'create', 'update', 'delete', 'link'
+    status VARCHAR(50) NOT NULL,      -- 'pending', 'synced', 'failed'
+    error_message TEXT,
+    created_at TIMESTAMP WITH TIME ZONE DEFAULT NOW(),
+    synced_at TIMESTAMP WITH TIME ZONE
+);
+
+-- =============================================================================
+-- Indexes
+-- =============================================================================
+
+-- Auth indexes
+CREATE INDEX IF NOT EXISTS idx_tokens_user ON engram.tokens(user_id);
+CREATE INDEX IF NOT EXISTS idx_tokens_hash ON engram.tokens(token_hash);
+CREATE INDEX IF NOT EXISTS idx_token_usage_token ON engram.token_usage(token_id);
+CREATE INDEX IF NOT EXISTS idx_token_usage_created ON engram.token_usage(created_at DESC);
+
+-- Memory indexes
+CREATE INDEX IF NOT EXISTS idx_memory_metadata_user ON engram.memory_metadata(user_id);
+CREATE INDEX IF NOT EXISTS idx_memory_metadata_project ON engram.memory_metadata(project_id);
+CREATE INDEX IF NOT EXISTS idx_memory_metadata_importance ON engram.memory_metadata(importance DESC);
+CREATE INDEX IF NOT EXISTS idx_memory_chunks_parent ON engram.memory_chunks(parent_memory_id);
+
+-- Graph sync indexes
+CREATE INDEX IF NOT EXISTS idx_graph_sync_memory ON engram.graph_sync_log(memory_id);
+CREATE INDEX IF NOT EXISTS idx_graph_sync_status ON engram.graph_sync_log(status);
+CREATE INDEX IF NOT EXISTS idx_graph_sync_pending ON engram.graph_sync_log(status) WHERE status = 'pending';
+
+-- =============================================================================
+-- Default User (for local development)
+-- =============================================================================
+
+INSERT INTO engram.users (email, name)
+VALUES ('local@squads.dev', 'Local Development User')
+ON CONFLICT (email) DO NOTHING;
+
+-- Grant permissions
+GRANT ALL PRIVILEGES ON SCHEMA engram TO squads;
+GRANT ALL PRIVILEGES ON ALL TABLES IN SCHEMA engram TO squads;
+GRANT USAGE, SELECT ON ALL SEQUENCES IN SCHEMA engram TO squads;
+
+-- =============================================================================
+-- Verify setup
+-- =============================================================================
+
+DO $$
+BEGIN
+    RAISE NOTICE 'Engram database initialized successfully';
+    RAISE NOTICE 'Extensions: vector, pg_trgm';
+    RAISE NOTICE 'Schema: engram';
+END $$;

package/docker/init-langfuse-db.sh

@@ -0,0 +1,10 @@
+#!/bin/bash
+# Create separate database for Langfuse to avoid schema conflicts
+set -e
+
+psql -v ON_ERROR_STOP=1 --username "$POSTGRES_USER" --dbname "$POSTGRES_DB" <<-EOSQL
+    CREATE DATABASE langfuse;
+    GRANT ALL PRIVILEGES ON DATABASE langfuse TO squads;
+EOSQL
+
+echo "Created langfuse database"

package/docker/otel-collector.yaml

@@ -0,0 +1,34 @@
+receivers:
+  otlp:
+    protocols:
+      http:
+        endpoint: 0.0.0.0:4318
+
+processors:
+  batch:
+    timeout: 5s
+    send_batch_size: 100
+
+exporters:
+  # Forward to Squads bridge (postgres primary, Langfuse optional)
+  otlphttp:
+    endpoint: http://squads-bridge:8080
+    encoding: json
+    headers: {}
+    tls:
+      insecure: true
+
+  # Debug logging
+  debug:
+    verbosity: detailed
+
+service:
+  pipelines:
+    metrics:
+      receivers: [otlp]
+      processors: [batch]
+      exporters: [otlphttp, debug]
+    logs:
+      receivers: [otlp]
+      processors: [batch]
+      exporters: [otlphttp, debug]

package/docker/squads-bridge/Dockerfile

@@ -0,0 +1,14 @@
+FROM python:3.11-slim
+
+ENV PYTHONUNBUFFERED=1
+
+WORKDIR /app
+
+COPY requirements.txt .
+RUN pip install --no-cache-dir -r requirements.txt
+
+COPY squads_bridge.py .
+
+EXPOSE 8080
+
+CMD ["python", "squads_bridge.py"]

package/docker/squads-bridge/Dockerfile.proxy

@@ -0,0 +1,14 @@
+FROM python:3.11-slim
+
+ENV PYTHONUNBUFFERED=1
+
+WORKDIR /app
+
+COPY requirements.txt .
+RUN pip install --no-cache-dir -r requirements.txt
+
+COPY anthropic_proxy.py .
+
+EXPOSE 8089
+
+CMD ["python", "anthropic_proxy.py"]