By ehartye
Track agent team task assignment, messages, and tool use across Claude Code sessions
This skill should be used when the user runs "/stalker-config" or asks to "configure agent-stalker", "set capture rules", "change what stalker records", "pause tracking for this project", "resume tracking", or "show the stalker config". Manages agent-stalker's content-capture rules and per-path pause state.
This skill should be used when the user runs "/stalker-tokens" or asks about token usage — "count tokens", "how many tokens did this session use", "show token usage", "token usage by agent", "tokens by model", "which session burned the most tokens", or "estimate the tokens in this file/text". Reports the real token usage agent-stalker captured, and counts/estimates tokens for arbitrary text or files for the Claude model family.
Analyze agent-stalker sessions that were flagged for triage in the dashboard. Use when the user runs /stalker-triage, asks to triage flagged sessions, or wants a pain analysis of sessions flagged via the "Flag for triage" button in the agent-stalker Insights dashboard. Reads each flagged session's digest, scores its workflow pain, and writes the result back so the dashboard displays it — no API key required.
This skill should be used when the user runs "/stalker-ui" or asks to "start the agent-stalker dashboard", "open the stalker UI", "launch the tracking dashboard", "view tracked sessions in the browser", or "stop the stalker server". Starts (or stops) the agent-stalker web dashboard.
This skill should be used when the user runs "/stalker" or asks to query agent-stalker's tracked data — "list tracked sessions", "show tool usage", "what events happened in session X", "list agents", "show tasks", "agent-stalker stats", or "query the stalker database". Runs the agent-stalker query engine over the captured SQLite database and presents the results.
Matches all tools
Hooks run on every tool call, not just specific ones
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.

Real-time observability for Claude Code agent teams. Track sessions, tool use, task lifecycle, and inter-agent messaging through a live web dashboard and CLI query engine.
Agent Stalker is a purely observational Claude Code plugin. It hooks into Claude's event system to record what agents do — without ever modifying or injecting into agent communication. Everything you see in the dashboard is a passive reflection of actual agent activity.
Track everything your agents do:
Full dashboard with kanban task board, agent/tool/event filter chips, and real-time activity feed. Each agent is color-coded for easy visual tracking.

Click any task card to see metadata, ownership, team assignment, and full event history showing every state transition.

Expand any activity event to see the raw tool input/output JSON data with syntax highlighting.

Add the marketplace to your Claude Code settings, then install the plugin:
# Add the marketplace (one-time setup)
claude plugins add-marketplace ehartye/hartye-claude-plugins
# Install agent-stalker
claude plugins install agent-stalker
To update to the latest version:
claude plugins update
Clone the repo and register it as a local marketplace:
git clone https://github.com/ehartye/agent-stalker.git
cd agent-stalker
bun install
Add to your Claude Code settings (~/.claude/settings.json):
{
"extraKnownMarketplaces": [
{
"name": "agent-stalker-dev",
"source": "directory",
"path": "/path/to/agent-stalker"
}
]
}
Then install:
claude plugins install agent-stalker --marketplace agent-stalker-dev
/stalker-ui
Opens the web dashboard at http://localhost:3141. Use --port <number> to change the port.
/stalker-ui stop
Stops the running dashboard server.
/stalker sessions # List recent sessions
/stalker tools --session <id> # Tool use frequency
/stalker events --since 1h # Recent events
/stalker tasks --status pending # Filter tasks by status
/stalker agents # List all spawned agents
/stalker stats # Summary statistics
/stalker tokens --by model # Captured token usage
/stalker-tokens --by session # Real captured token usage (totals + breakdown)
/stalker-tokens --since 24h # Usage in a recent window
/stalker-tokens count <file> # Tokenize a file (exact with ANTHROPIC_API_KEY, else an estimate)
Real counts come from the usage table (parsed from transcripts on SessionEnd). Ad-hoc tokenization of arbitrary text/files is exact when ANTHROPIC_API_KEY is set (Anthropic's free count_tokens API), otherwise a clearly-labelled local estimate.
Note: the commands above are now skills (
/stalker,/stalker-ui,/stalker-config,/stalker-tokens,/stalker-triage) — same invocation, plus contextual auto-triggering.
/stalker-config # Show current config
/stalker-config set Bash full # Capture full Bash output
/stalker-config set Read metadata # Metadata only for Read
/stalker-config reset # Restore defaults
Default capture rules:
| Tool | Rule |
|---|---|
| Edit, Write | metadata |
| Read, Glob, Grep | metadata |
| Bash | maxLength 2000 |
| Everything else | maxLength 500 |
What is captured: events are stored in plaintext SQLite at
~/.claude/agent-stalker.db. By default file contents are not stored (metadata only); opt into full capture per-tool with/stalker-config set Edit full. The dashboard binds127.0.0.1— to view it from another machine, set"ui": { "host": "0.0.0.0" }in~/.claude/agent-stalker.config.json. If browsing by machine name rather than IP, also add the name to"allowedHosts"(e.g.["office-pc"]).
npx claudepluginhub ehartye/agent-stalker --plugin agent-stalkerTurn markdown documents (with embedded mermaid diagrams) into themed, searchable, paged PDFs and Microsoft Word DOCX files via WeasyPrint and python-docx. Bundles six skills (publish, preprocess, theme-advisor, theme-gallery, install-fonts, rebuild-themes) plus eight built-in themes (atlas/phosphor/arcade/signal in light + dark).
Find and integrate free 2D game assets into Phaser JS projects. Searches trusted free sources (Kenney, OpenGameArt, itch.io), presents a visual preview UI for selection, downloads assets into your project, and generates ready-to-use Phaser loading code.
Salesforce browser automation via SF CLI authentication - control browser sessions, navigate Lightning pages, fill forms, and manage records
Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques
Mission Control for Claude Code — auto-starts a web dashboard, provides 90 MCP tools (sessions, stats, live monitoring, projects, teams, insights, coaching, and more), and adds 9 skills including /session-recap, /daily-cost, /standup, /coaching, /insights, and /team-status.
A plugin for Claude Code agent team workflows — session viewing, analysis, and best practice evaluation
Agent dashboard TUI — tracks AI agent state via hooks for the fleet CLI
Real-time observability dashboard for Claude Code agents
Observability platform for Claude Code and Agent SDK — monitor, debug, and control AI coding agents
Multi-agent tmux dashboard with skills, hooks, and workflow automation