devteam-status
npx machina-cli add skill michael-harris/devteam/devteam-status --openclawSession state: !source "${CLAUDE_PLUGIN_ROOT}/scripts/state.sh" && get_current_session 2>/dev/null || echo "No active session"
Recent events: !sqlite3 .devteam/devteam.db "SELECT type, message FROM events ORDER BY created_at DESC LIMIT 10" 2>/dev/null || echo "No events"
DevTeam Status Command
Command: /devteam:status [options]
Display system health, current progress, costs, and execution history.
Usage
# Current session status
/devteam:status
# Show execution history
/devteam:status --history
/devteam:status --history 20 # Last 20 sessions
# Cost breakdown
/devteam:status --costs
/devteam:status --costs --detailed
# Agent performance metrics
/devteam:status --agents
# Check specific session
/devteam:status --session session-20260129-123456-abc123
# Show all (comprehensive report)
/devteam:status --all
Options
| Option | Description |
|---|---|
--history [n] | Show last n sessions (default: 10) |
--costs | Show cost breakdown |
--detailed | Show detailed breakdown (with --costs) |
--agents | Show agent performance metrics |
--session <id> | Show specific session details |
--all | Show comprehensive report |
--json | Output as JSON |
Your Process
Check for Running Session
source "${CLAUDE_PLUGIN_ROOT}/scripts/state.sh"
if is_session_running; then
show_current_session_status
else
show_system_overview
fi
Output: Current Session
DevTeam Status
Session: session-20260129-103045-a1b2c3
Started: 2026-01-29 10:30:45 (23 min ago)
Command: /devteam:implement --sprint 1
Progress
Sprint: SPRINT-001 (Add user authentication)
Tasks: 60% (3/5 complete)
Current Task: TASK-004 - Implement JWT middleware
Agent: backend:api-developer-typescript
Model: sonnet (escalated from haiku)
Iteration: 3 of 10
Model Usage
Model | Runs | Success | Tokens | Cost
haiku | 12 | 83% | 45,231 | $0.06
sonnet | 4 | 75% | 28,102 | $0.42
opus | 1 | 100% | 15,883 | $2.38
Total | 17 | 82% | 89,216 | $2.86
Quality Gates (Last Iteration)
Lint passed
Type Check passed
Tests 2 failing (AuthMiddleware.test.ts)
Security pending
Coverage pending
Escalations
10:35 haiku -> sonnet 2 test failures (api_developer)
10:48 sonnet -> opus Architecture issue detected
Total: 2 escalations
Circuit Breaker
State: CLOSED (healthy)
Consecutive Failures: 1 of 5
Bug Council: Not activated
Output: No Active Session
DevTeam Status
Status: Idle (no active session)
Recent Sessions
Session | Command | Status | Cost
session-20260129-093012-xxx | /devteam:bug | done | $1.24
session-20260128-161532-xxx | /devteam:impl... | done | $4.82
session-20260128-142201-xxx | /devteam:impl... | fail | $2.11
session-20260128-103045-xxx | /devteam:plan | done | $0.38
Active Plans
# | Plan | Status | Progress
1 | user-authentication | in_progress | 80%
2 | dark-mode-feature | ready | 0%
Lifetime Statistics
Total Sessions: 47
Success Rate: 89%
Total Cost: $127.43
Avg Cost/Session: $2.71
Most Used Agents:
1. backend:api-developer-typescript (142 runs, 91% success)
2. quality:test-writer (98 runs, 87% success)
3. frontend:developer (76 runs, 88% success)
Bug Council Activations: 3
Avg Escalations/Session: 1.2
Tip: Use /devteam:implement to start a new session
Output: Cost Details (--costs --detailed)
DevTeam Cost Report
Period: Last 30 days
Cost by Model
haiku $12.43 (10%)
sonnet $68.21 (54%)
opus $46.79 (36%)
Total: $127.43
Cost by Command Type
implement $89.21 (70%)
bug $24.33 (19%)
plan $10.12 (8%)
issue $3.77 (3%)
Cost by Agent (Top 10)
Agent | Runs | Cost | Avg/Run
backend:api-developer-typescript | 142 | $34.21 | $0.24
quality:test-writer | 98 | $18.43 | $0.19
quality:security-auditor | 23 | $15.82 | $0.69
frontend:developer | 76 | $12.11 | $0.16
bug_council (combined) | 3 | $11.24 | $3.75
...
Eco Mode Savings
Sessions with --eco: 12
Estimated savings: $18.32 (23% reduction)
Recommendation: Use --eco for routine tasks to save costs
Output: Agent Performance (--agents)
Agent Performance Report
Agent Success Rates
Agent | Runs | Success | Avg Time
backend:api-developer-typescript | 142 | 91% | 2.3 min
backend:api-developer-python | 67 | 89% | 2.1 min
frontend:developer | 76 | 88% | 3.1 min
quality:test-writer | 98 | 87% | 1.8 min
quality:security-auditor | 23 | 96% | 4.2 min
diagnosis:root-cause-analyst | 15 | 80% | 3.5 min
Common Failure Reasons
Reason | Count | Usually Resolved By
Test failures | 34 | Model escalation
Type errors | 21 | Same model retry
Lint errors | 12 | Same model retry
Integration issues | 8 | Bug Council
Scope violation | 5 | Scope clarification
Escalation Patterns
haiku -> sonnet: 67 times (avg after 1.8 failures)
sonnet -> opus: 23 times (avg after 2.1 failures)
opus -> council: 3 times (avg after 2.7 failures)
Most escalated tasks:
- Security fixes (78% escalation rate)
- Architecture changes (65% escalation rate)
- Complex integrations (52% escalation rate)
Implementation
Query Current Session
-- Get current session with summary
SELECT * FROM v_session_summary WHERE status = 'running';
-- Get model usage for current session
SELECT * FROM v_model_usage WHERE session_id = (
SELECT id FROM sessions WHERE status = 'running'
);
-- Get recent events
SELECT timestamp, event_type, message
FROM events
WHERE session_id = (SELECT id FROM sessions WHERE status = 'running')
ORDER BY timestamp DESC
LIMIT 20;
-- Get gate results for current iteration
SELECT gate, passed, error_count
FROM gate_results
WHERE session_id = (SELECT id FROM sessions WHERE status = 'running')
AND iteration = (SELECT current_iteration FROM sessions WHERE status = 'running');
Query History
-- Recent sessions
SELECT
id,
command,
status,
started_at,
ended_at,
ROUND(total_cost_cents / 100.0, 2) as cost_dollars
FROM sessions
ORDER BY started_at DESC
LIMIT 10;
-- Lifetime statistics
SELECT
COUNT(*) as total_sessions,
ROUND(AVG(CASE WHEN status = 'completed' THEN 1.0 ELSE 0.0 END) * 100, 1) as success_rate,
SUM(total_cost_cents) / 100.0 as total_cost
FROM sessions;
Query Agent Performance
-- Agent performance
SELECT * FROM v_agent_performance ORDER BY total_runs DESC;
-- Gate pass rates
SELECT * FROM v_gate_pass_rates;
See Also
/devteam:list- List plans and tasks/devteam:reset- Reset stuck state/devteam:implement- Start implementation
Source
git clone https://github.com/michael-harris/devteam/blob/main/skills/devteam-status/SKILL.mdView on GitHub Overview
DevTeam Status provides a consolidated view of the system's health, current sprint progress, cost breakdowns, and recent execution history. It helps operators spot bottlenecks, cost spikes, and activity trends at a glance.
How This Skill Works
The /devteam:status command queries the current session state and recent events, and may query a local SQLite database for history. It formats a report based on flags like --history, --costs, --agents, --session, --all, and --json, delivering either a human-friendly report or JSON for automation.
When to Use It
- During live operation to monitor system health and sprint progress
- After a run to review recent execution history
- Before a release to audit cost breakdowns
- When debugging agent performance or bottlenecks with --agents
- When you need a comprehensive snapshot of a specific session or a full report with --all
Quick Start
- Step 1: Run /devteam:status
- Step 2: Add flags such as --history, --costs, or --agents
- Step 3: For deep inspection, specify --session <id> or --all; use --json for automation
Best Practices
- Run with --history to compare recent sessions and detect drift
- Combine --costs (and --detailed) to understand budgeting and spend
- Include --agents to identify underperforming agents
- Use --session <id> to drill into a single session for diagnosis
- Add --json for easy integration with dashboards or CI
Example Use Cases
- /devteam:status to view current status
- /devteam:status --history 20 to show last 20 sessions
- /devteam:status --costs --detailed for a deep cost breakdown
- /devteam:status --agents to inspect agent performance metrics
- /devteam:status --session session-20260129-123456-abc123 --all to inspect a specific session with full report