Monitoring plugins

Energy meter for Claude Code — watt-hours, kettle-boils, and an optional cyberpunk city.

2 commands

Real-time statusline showing context usage, active tools, running agents, and todo progress. Always visible below your input, zero config required.

2 commands

Statusline modular em bash, com previsão de estouro de rate limit, quota de provedor corporativo e saúde de sprint. Widgets carregados sob demanda a partir de um único config.json.

1 command

Status line under the prompt (context bar, model, effort, folder and branch, spend limit), live task list and next-step card, context details, model and effort switchers, and image previews.

1 command

Access PostHog analytics, feature flags, experiments, error tracking, and insights directly from Claude Code.

81 skills · 3 commands · 1 agent · 1 MCP server · hooks

Use Amplitude as an expert analyst — instrument Amplitude, discover product opportunities, analyze charts, create dashboards, manage experiments, and understand users and accounts.

37 skills · 1 MCP server

Use Amplitude as an expert analyst — instrument Amplitude, discover product opportunities, analyze charts, create dashboards, manage experiments, and understand users and accounts.

37 skills · 1 MCP server

Skills, agents, and workflows for Honeycomb observability — query patterns, production investigations, SLOs, OpenTelemetry instrumentation, and Beeline migration. Designed to complement the Honeycomb MCP server.

11 skills · 1 command · 2 agents · 1 MCP server · hooks

Add Logfire observability to Python applications with auto-instrumentation for FastAPI, httpx, asyncpg, SQLAlchemy, and more

6 skills · 4 commands · 1 MCP server

Add Logfire observability to Python applications with auto-instrumentation for FastAPI, httpx, asyncpg, SQLAlchemy, and more

6 skills · 4 commands · 1 MCP server

Eight skills (status, incidents, investigate, environments, tasks, rca, changes, knowledge), a read-only investigator agent, and the hosted MCP server with 25 tools, 4 prompts and 5 resources.

8 skills · 1 agent · 1 MCP server

Connect Claude Code to Brizz — instrument an AI agent, then investigate its sessions, issues, and cost. Bundles the Brizz MCP server and the `brizz` CLI skills.

3 skills · 1 MCP server

Local web dashboard for Claude Code — monitor rate limits, manage agents/skills/hooks/MCP servers, browse session history, track daily activity and usage rankings

4 commands

百炼 Coding Plan 用量监控，显示近5小时/周/月用量百分比，与其他 statusLine 插件并存

3 commands

Traces Claude Code sessions and tool calls as OpenTelemetry spans in Sentry.

1 skill · hooks

Tracks Anthropic API usage (5h/7d rolling limits) and automatically blocks opus sub-agent calls when thresholds are exceeded, forcing sonnet fallback. Includes real-time usage display on every prompt.

1 command · hooks

Statusline HUD with a per-position gradient progress bar (cool blue → warm red). Each dot shows its own color; the warmer colors only appear when the bar is near full, so tension is visible at a glance.

2 commands

Statusline HUD for DeepSeek, Kimi, GLM, MiniMax, and MiMo: context bar, provider quotas/balance, token cost estimation (¥), and git status. Run /multi-hud:setup after install.

2 commands

Real-time statusline showing context usage, active tools, running agents, and todo progress. Always visible below your input, zero config required.

1 command

Per-turn token usage telemetry for Claude Code. Ships usage metadata to your organisation's self-hosted receiver — no message content is ever collected.

Grafana dashboards, Loki LogQL, variable templating, alerting, and API queries

Stream recent logs from systemd journal and application log files

Grafana alerts, Loki LogQL, SLO dashboards, Telegram notifications

See your LiteLLM virtual key inside Claude Code: budget, spend, limits, expiry, models and recent usage. Admins can also create keys, grant extra budget and read router fallbacks.

See and steer your subagents: cards with cost, tokens, context and progress, tree and team views, spawn caps, permission waits, idle alerts and a Stop button.

A token budget for the session and each subagent that warns, refuses new subagents or stops tool calls, and a prompt-cache timer with an opt-in keep-warm.

Per-session token tally. Records usage totals in the store and prints the turn total; it is not a token stream or a server-side audit chain.

Observable audit trace only. Records tool calls and turn completions to a capped store log; it never blocks work.

Access PostHog analytics, feature flags, experiments, error tracking, and insights directly from Claude Code.

161 skills · 3 commands · 1 agent · 1 MCP server · hooks

Access PostHog analytics, feature flags, experiments, error tracking, and insights directly from Claude Code.

161 skills · 3 commands · 1 agent · 1 MCP server · hooks