By Bolivar1234
Activity tracker for Claude Code — tracks sessions, decisions, and progress across projects with a local dashboard
Health check for Claude Pulse — verify dependencies, database, hooks, and recent tracking activity
Query the insights/brain database — search decisions, progress, blockers, patterns across all projects. Use when the user asks "what decisions did I make", "show me blockers", "what patterns have I noticed", "search my insights for X", or wants to find a past decision.
Show the latest work across all projects — recent sessions, what was accomplished, and what's blocked. Use when the user asks "what did I work on", "what's the latest", "show recent activity", "where did I leave off", or wants a cross-project summary.
List all tracked projects with session counts, last activity, and recent insights. Use when the user asks "what projects am I working on", "show my projects", "list projects", or wants cross-project context.
Show Claude Pulse tracking status — session counts, events, lines written, database health
Executes bash commands
Hook triggers when Bash tool is used
Modifies files
Hook triggers on file write and edit operations
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Executables (bin/) — files in this plugin's bin/directory are added to the Bash tool's PATH while the plugin is enabled.
One command. Global memory for Claude Code.
npx claude-pulse init
Claude Pulse captures every session, every decision, and every line of code across all your projects — automatically, locally, and queryable anytime. It gives Claude Code a persistent memory and gives you a dashboard to see your work.
No cloud. No accounts. No data leaves your machine.
| Feature | How it works |
|---|---|
| Session tracking | Every start, stop, and tool call recorded automatically |
| Code metrics | Lines added/removed per file, per session, per day |
| Brain layer | Structured progress, decisions, and blockers captured at session end |
| Smart gating | Only asks for summaries on meaningful sessions — trivial ones auto-close |
| Cross-project context | Claude knows what you did in other projects when you start a session |
| Knowledge capture | Store patterns, fixes, and context mid-session with /pulse-remember |
| Audit trail | User identity, diff capture, and exportable records for compliance |
| Dashboard | Visual overview at localhost:3141 — activity, brain, projects, timeline |
# macOS
brew install jq sqlite3
# Ubuntu/Debian
sudo apt install jq sqlite3
git clone https://github.com/Clemens865/Claude-Pulse.git
cd Claude-Pulse
npm install
npx claude-pulse init
That's it. Claude Pulse is now active for all projects. Use Claude Code normally — everything is captured in the background.
npx claude-pulse start
Open http://localhost:3141.
Key numbers at a glance: total sessions, net lines written, hours spent, active projects, streaks. Daily activity for the last 14 days. Top tools, skills, and frameworks.
A searchable timeline of your reasoning — not just your code. Structured entries for progress, decisions, blockers, patterns, and fixes. Filterable by type and project. This is the log of your thinking across sessions.
All projects ranked by activity. Click any project for session history, most-edited files, daily breakdowns.
90-day GitHub-style heatmap. Darker squares = more active days.
Full event-by-event replay of any session. Every edit shows an expandable diff view. User identity (who@hostname) displayed when available.
Database info, record counts, and an export section with project/date filtering. Download your data as JSON or CSV anytime.
| Command | What it does |
|---|---|
npx claude-pulse init | Set up hooks and database (safe to re-run) |
npx claude-pulse start | Open the dashboard |
npx claude-pulse status | Quick terminal summary |
npx claude-pulse doctor | Health check — verify everything works |
npx claude-pulse export | Export data as JSON, CSV, or NDJSON |
npx claude-pulse verify | Audit integrity check |
npx claude-pulse uninstall | Remove hooks (data preserved) |
Export supports filtering:
claude-pulse export --format json --start 2026-03-01 --end 2026-03-31
claude-pulse export --format csv --table insights
These work globally — any project, any directory:
| Command | What it does |
|---|---|
/pulse-projects | List all tracked projects with stats |
/pulse-projects my-app | Details and insights for a specific project |
/pulse-latest | Last 3 days of work across all projects |
/pulse-latest 7 | Last 7 days |
/pulse-insights | Overview of all insights |
/pulse-insights type:decision | All decisions you've made |
/pulse-insights type:blocked | All current blockers |
/pulse-insights auth | Search insights by keyword |
Some knowledge emerges mid-session. /pulse-remember captures it:
/pulse-remember fix: sqlite3 needs absolute paths on macOS
/pulse-remember pattern: all API routes use getDb() singleton
/pulse-remember context: deploy freeze next week for mobile release
Claude distills your input into a concise summary and stores both layers:
These show up as "Knowledge" at your next session start — before yesterday's progress, because accumulated knowledge is more valuable.
Claude Pulse captures a complete activity trail suitable for AI code audits:
Enterprise AI agent orchestration plugin with 150+ commands, 74+ specialized agents, SPARC methodology, swarm coordination, GitHub integration, and neural training capabilities
Lazy senior dev mode. Forces the simplest, shortest solution that actually works: YAGNI, stdlib first, no unrequested abstractions.
npx claudepluginhub bolivar1234/claude-pulseConsult multiple AI coding agents (Gemini, OpenAI, Grok, Perplexity, plus codex, antigravity, and grok CLIs when installed) to get diverse perspectives on coding problems
v9.52.0 - Reliability wave: tangle contextual review correction loop with hard round ceiling, progress-supervised review rounds (per-agent stall watch, descendant-tree kills), council diversity and agy pin fixes, marketplace generator source-of-truth fix, provider troubleshooting runbook and cost-expectations docs. Run /octo:setup.
Permanent coding companion for Claude Code — survives any update. MCP-based terminal pet with ASCII art, stats, reactions, and personality.