claudefleet-agent 0.1.2

ClaudeFleet local agent — scans Claude Code usage and syncs to the central platform

ClaudeFleet Agent — Install & Scan Guide

The agent scans Claude Code's local transcript files, stores usage in a local SQLite database, and (optionally) syncs it to the central server. Scanning works fully offline — no server required.

Tracked activity ≠ official quota. All numbers are token counts parsed from local transcripts — an estimate, not your Claude Max/Pro billing or quota.

---

1. Prerequisites

• Python 3.10+ (check with python --version)
• Claude Code installed and used at least once on this machine, so transcripts exist under ~/.claude/projects/ (Windows: %USERPROFILE%\.claude\projects\).

The agent uses only the Python standard library — there is nothing to pip install for scanning.

For central mode (sending usage to the dashboard) you install the claudefleet command from the repo, which needs git on PATH — see §7.

---

2. Install

Clone the repository and move into the agent folder:

git clone <your-repo-url> claudefleet
cd claudefleet\agent

You can run it three ways:

A. Install from PyPI (public release) — once published, users can install:

pip install claudefleet-agent

B. No install (simplest) — run it as a module from the agent/ folder:

python -m claudefleet --help

C. Install the claudefleet command locally (so you can run it from anywhere):

pip install -e .
claudefleet --help

Both install options are equivalent for command usage; the rest of this guide uses python -m claudefleet.

The package is now public-ready as claudefleet-agent. If the package has already been published on PyPI, use the pip install claudefleet-agent command above.

---

3. Use the agent as a Python SDK

The package now exposes a simple SDK interface via claudefleet.Agent.

Install the package

pip install -e .

Example usage

from claudefleet import Agent

agent = Agent(display_name="PC-01")
agent.register(
    server_url="http://127.0.0.1:8000",
    api_key="cfk_...",
    display_name="PC-01",
)
print(agent.scan())
print(agent.sync())
print(agent.health())

The SDK also supports running the daemon programmatically:

agent = Agent(display_name="PC-01")
agent.daemon()

---

4. Run a scan

Give this machine a name (once), then scan:

python -m claudefleet identity --display-name PC-01
python -m claudefleet scan

Example output:

  [NEW] C:\Users\you\.claude\projects\my-proj\<uuid>.jsonl  (+266 events)
  ...
scan complete: new=18 updated=0 skipped=0 events+=1787

The scan is incremental and idempotent:

• Unchanged files are skipped; changed files are read only from where they left off.
• Running it again processes nothing new (events+=0) — it never double-counts.

Run scan whenever you want fresh data (or schedule it — see §6).

---

4. View your usage

python -m claudefleet today     # today's tokens by model
python -m claudefleet week      # last 7 days
python -m claudefleet stats     # all-time totals, by model, top projects

stats example:

  ClaudeFleet - all-time tracked usage
  Sessions:       13
  Input tokens:   6.2M   Output tokens: 1.1M   Cache read: 280M ...
  By model:  claude-opus-4-8  ...  claude-sonnet-5 ...
  Top projects:  Github/hotel-demo  ...
  (estimate only - not official Max/Pro quota)

---

5. Where your data lives

What	Default location	Override with
Usage database	~/.claude/claudefleet/usage.db	CLAUDEFLEET_DB
Machine identity/config	~/.claude/claudefleet/agent.json	CLAUDEFLEET_CONFIG
Transcripts it reads (read-only)	~/.claude/projects/**/*.jsonl	auto-discovered

Example with a custom DB path (PowerShell):

$env:CLAUDEFLEET_DB = "D:\data\usage.db"
python -m claudefleet scan

The agent never modifies Claude Code's files and never stores prompts, responses, or source code — only token counts and metadata.

---

6. Automatic scanning (Windows Task Scheduler)

Run a scan every 15 minutes without thinking about it (one line):

schtasks /Create /SC MINUTE /MO 15 /TN "ClaudeFleet Scan" /TR "python -m claudefleet scan --quiet" /ST 00:00

If you installed via the "Connect PC" flow (§7) or deploy/install.ps1, the task is instead named "ClaudeFleet Scan+Sync" and also pushes data to the central server every 15 minutes.

Stop the agent from scanning

Remove whichever scheduled task applies to how you installed:

schtasks /Delete /TN "ClaudeFleet Scan" /F         # local scan-only task
schtasks /Delete /TN "ClaudeFleet Scan+Sync" /F    # Connect PC / install.ps1 task

To pause it instead of deleting it (keeps run history, easy to re-enable):

schtasks /Change /TN "ClaudeFleet Scan+Sync" /DISABLE
schtasks /Change /TN "ClaudeFleet Scan+Sync" /ENABLE   # resume later

schtasks /TR does not go through cmd.exe, so a raw scan && sync command line will not chain correctly — it's run via a small claudefleet-scan-sync.cmd wrapper batch file instead. If you set the task up by hand, point /TR at a .cmd wrapper rather than an inline &&.

---

7. Send data to the central server (optional)

Local scanning is enough for one machine. To feed a central dashboard:

1. Sign in to the web app and open Connect PC (or ask your admin for an API key).
2. The web app generates a one-line setup command for the machine you want to track. This is the command you run in PowerShell on that PC.
3. The command installs the agent, registers the machine with the server, scans local Claude Code transcripts, and syncs the results back to the dashboard.

If you only have the website link, that is enough. The site does not scan your PC from the browser; it only generates the install/connect command and gives you the server URL and API key to use.

pip install claudefleet-agent
claudefleet register --server https://YOUR-API-URL --api-key cfk_... --display-name PC-01
claudefleet scan
claudefleet sync

Or use the Windows installer from the repo (deploy/install.ps1), which does all four steps above and sets up the recurring scan+sync task automatically.

--server must be the API server (e.g. http://localhost:8000 in dev), not the dashboard's frontend URL (http://localhost:5173/...). Pointing it at the frontend URL will fail to register.

The agent connects by calling the server at the given API URL and authenticating with the supplied API key. After registration, it reads local transcript files, aggregates token events, and sends only usage metadata to the dashboard.

An administrator can also create an API key under Admin → Agent API keys, then share the key and API URL with each user.

sync only sends events the server hasn't seen; if the server is down it simply retries next time (nothing is lost, nothing is double-counted). The dashboard shows a system as "Never synced" until the first successful sync call — registering alone, or running scan without sync, is not enough.

---

8. Troubleshooting

Symptom	Fix
scan complete: new=0 ... events+=0 on first run	No transcripts found. Confirm %USERPROFILE%\.claude\projects\ exists and you've used Claude Code.
python not found	Install Python 3.10+ and ensure it's on PATH (py -3 also works on Windows).
Want a clean re-scan	Delete the usage DB (%USERPROFILE%\.claude\claudefleet\usage.db) and run scan again.
sync says "Central mode not configured"	Run register first with --server and --api-key.
sync fails / offline	Expected when the server is unreachable; it retries on the next run.

---

Command reference

python -m claudefleet scan       [--display-name NAME] [--quiet]
python -m claudefleet today | week | stats
python -m claudefleet identity   [--display-name NAME] [--set-display-name NAME]
python -m claudefleet register   --server URL --api-key KEY [--display-name NAME]
python -m claudefleet sync       [--quiet]
python -m claudefleet heartbeat
python -m claudefleet --version

Run the test suite with pip install pytest && python -m pytest (33 tests).