clawmeter 0.7.4__tar.gz

clawmeter-0.7.4/.dockerignore

@@ -0,0 +1,10 @@
+.github
+__pycache__
+*.egg-info
+*.pyc
+.pytest_cache
+tests/
+docs/
+.venv
+*.md
+!README.md

clawmeter-0.7.4/.github/workflows/ci.yml

@@ -0,0 +1,30 @@
+name: CI
+
+on:
+  push:
+    branches: [main]
+  pull_request:
+    branches: [main]
+
+jobs:
+  test:
+    runs-on: ubuntu-latest
+    strategy:
+      matrix:
+        python-version: ["3.10", "3.12", "3.13"]
+
+    steps:
+      - uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+
+      - uses: astral-sh/setup-uv@v4
+
+      - name: Set up Python ${{ matrix.python-version }}
+        run: uv python install ${{ matrix.python-version }}
+
+      - name: Install dependencies
+        run: uv sync --group dev
+
+      - name: Run tests
+        run: uv run pytest -v

clawmeter-0.7.4/.github/workflows/publish.yml

@@ -0,0 +1,148 @@
+# ============================================
+# CLAWMETER - Publish to PyPI
+# ============================================
+# Triggered on version tags (v*) or manual dispatch
+# Builds and publishes to PyPI via trusted publishing
+# ============================================
+
+name: Publish
+
+on:
+  push:
+    tags:
+      - 'v*'
+  workflow_dispatch:
+    inputs:
+      target:
+        description: 'Publish target'
+        required: true
+        default: 'testpypi'
+        type: choice
+        options:
+          - testpypi
+          - pypi
+
+env:
+  PYTHON_VERSION: "3.12"
+
+jobs:
+  # ============================================
+  # BUILD
+  # ============================================
+  build:
+    name: Build Distribution
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 0  # Full history for version detection
+
+      - name: Set up Python
+        uses: actions/setup-python@v5
+        with:
+          python-version: ${{ env.PYTHON_VERSION }}
+
+      - name: Install build tools
+        run: |
+          python -m pip install --upgrade pip
+          pip install build twine
+
+      - name: Determine version from tag
+        if: startsWith(github.ref, 'refs/tags/v')
+        run: echo "SETUPTOOLS_SCM_PRETEND_VERSION=${GITHUB_REF#refs/tags/v}" >> "$GITHUB_ENV"
+
+      - name: Build package
+        run: python -m build
+
+      - name: Check package
+        run: twine check dist/*
+
+      - name: List artifacts
+        run: ls -la dist/
+
+      - name: Upload build artifacts
+        uses: actions/upload-artifact@v4
+        with:
+          name: dist
+          path: dist/
+          retention-days: 30
+
+  # ============================================
+  # PUBLISH TO TEST PYPI
+  # ============================================
+  publish-testpypi:
+    name: Publish to TestPyPI
+    runs-on: ubuntu-latest
+    needs: build
+    if: |
+      (github.event_name == 'workflow_dispatch' && github.event.inputs.target == 'testpypi')
+    environment:
+      name: testpypi
+      url: https://test.pypi.org/project/clawmeter/
+    permissions:
+      id-token: write  # For trusted publishing
+
+    steps:
+      - name: Download artifacts
+        uses: actions/download-artifact@v4
+        with:
+          name: dist
+          path: dist/
+
+      - name: Publish to TestPyPI
+        uses: pypa/gh-action-pypi-publish@release/v1
+        with:
+          repository-url: https://test.pypi.org/legacy/
+
+  # ============================================
+  # PUBLISH TO PYPI
+  # ============================================
+  publish-pypi:
+    name: Publish to PyPI
+    runs-on: ubuntu-latest
+    needs: build
+    if: |
+      (github.event_name == 'workflow_dispatch' && github.event.inputs.target == 'pypi') ||
+      (github.event_name == 'push' && startsWith(github.ref, 'refs/tags/v'))
+    environment:
+      name: pypi
+      url: https://pypi.org/project/clawmeter/
+    permissions:
+      id-token: write  # For trusted publishing
+
+    steps:
+      - name: Download artifacts
+        uses: actions/download-artifact@v4
+        with:
+          name: dist
+          path: dist/
+
+      - name: Publish to PyPI
+        uses: pypa/gh-action-pypi-publish@release/v1
+
+  # ============================================
+  # POST-PUBLISH VERIFICATION
+  # ============================================
+  verify:
+    name: Verify Installation
+    runs-on: ubuntu-latest
+    needs: [publish-pypi]
+    if: github.event_name == 'push' && startsWith(github.ref, 'refs/tags/v')
+    steps:
+      - name: Wait for PyPI propagation
+        run: sleep 60
+
+      - name: Set up Python
+        uses: actions/setup-python@v5
+        with:
+          python-version: ${{ env.PYTHON_VERSION }}
+
+      - name: Install from PyPI
+        run: |
+          pip install clawmeter
+          clawmeter --version
+          clawmeter --help
+
+      - name: Verify import
+        run: python -c "import clawmeter; print(f'clawmeter {clawmeter.__version__}')"

clawmeter-0.7.4/.gitignore

@@ -0,0 +1,211 @@
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[codz]
+*$py.class
+
+# C extensions
+*.so
+
+# Distribution / packaging
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+share/python-wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+MANIFEST
+
+# PyInstaller
+#  Usually these files are written by a python script from a template
+#  before PyInstaller builds the exe, so as to inject date/other infos into it.
+*.manifest
+*.spec
+
+# Installer logs
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Unit test / coverage reports
+htmlcov/
+.tox/
+.nox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+*.cover
+*.py.cover
+.hypothesis/
+.pytest_cache/
+cover/
+
+# Translations
+*.mo
+*.pot
+
+# Django stuff:
+*.log
+local_settings.py
+db.sqlite3
+db.sqlite3-journal
+
+# Flask stuff:
+instance/
+.webassets-cache
+
+# Scrapy stuff:
+.scrapy
+
+# Sphinx documentation
+docs/_build/
+
+# PyBuilder
+.pybuilder/
+target/
+
+# Jupyter Notebook
+.ipynb_checkpoints
+
+# IPython
+profile_default/
+ipython_config.py
+
+# pyenv
+#   For a library or package, you might want to ignore these files since the code is
+#   intended to run in multiple environments; otherwise, check them in:
+# .python-version
+
+# pipenv
+#   According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control.
+#   However, in case of collaboration, if having platform-specific dependencies or dependencies
+#   having no cross-platform support, pipenv may install dependencies that don't work, or not
+#   install all needed dependencies.
+#Pipfile.lock
+
+# UV
+#   Similar to Pipfile.lock, it is generally recommended to include uv.lock in version control.
+#   This is especially recommended for binary packages to ensure reproducibility, and is more
+#   commonly ignored for libraries.
+#uv.lock
+
+# poetry
+#   Similar to Pipfile.lock, it is generally recommended to include poetry.lock in version control.
+#   This is especially recommended for binary packages to ensure reproducibility, and is more
+#   commonly ignored for libraries.
+#   https://python-poetry.org/docs/basic-usage/#commit-your-poetrylock-file-to-version-control
+#poetry.lock
+#poetry.toml
+
+# pdm
+#   Similar to Pipfile.lock, it is generally recommended to include pdm.lock in version control.
+#   pdm recommends including project-wide configuration in pdm.toml, but excluding .pdm-python.
+#   https://pdm-project.org/en/latest/usage/project/#working-with-version-control
+#pdm.lock
+#pdm.toml
+.pdm-python
+.pdm-build/
+
+# pixi
+#   Similar to Pipfile.lock, it is generally recommended to include pixi.lock in version control.
+#pixi.lock
+#   Pixi creates a virtual environment in the .pixi directory, just like venv module creates one
+#   in the .venv directory. It is recommended not to include this directory in version control.
+.pixi
+
+# PEP 582; used by e.g. github.com/David-OConnor/pyflow and github.com/pdm-project/pdm
+__pypackages__/
+
+# Celery stuff
+celerybeat-schedule
+celerybeat.pid
+
+# SageMath parsed files
+*.sage.py
+
+# Environments
+.env
+.envrc
+.venv
+env/
+venv/
+ENV/
+env.bak/
+venv.bak/
+
+# Spyder project settings
+.spyderproject
+.spyproject
+
+# Rope project settings
+.ropeproject
+
+# mkdocs documentation
+/site
+
+# mypy
+.mypy_cache/
+.dmypy.json
+dmypy.json
+
+# Pyre type checker
+.pyre/
+
+# pytype static type analyzer
+.pytype/
+
+# Cython debug symbols
+cython_debug/
+
+# PyCharm
+#  JetBrains specific template is maintained in a separate JetBrains.gitignore that can
+#  be found at https://github.com/github/gitignore/blob/main/Global/JetBrains.gitignore
+#  and can be added to the global gitignore or merged into this file.  For a more nuclear
+#  option (not recommended) you can uncomment the following to ignore the entire idea folder.
+#.idea/
+
+# Abstra
+# Abstra is an AI-powered process automation framework.
+# Ignore directories containing user credentials, local state, and settings.
+# Learn more at https://abstra.io/docs
+.abstra/
+
+# Visual Studio Code
+#  Visual Studio Code specific template is maintained in a separate VisualStudioCode.gitignore 
+#  that can be found at https://github.com/github/gitignore/blob/main/Global/VisualStudioCode.gitignore
+#  and can be added to the global gitignore or merged into this file. However, if you prefer, 
+#  you could uncomment the following to ignore the entire vscode folder
+# .vscode/
+
+# Ruff stuff:
+.ruff_cache/
+
+# PyPI configuration file
+.pypirc
+
+# Cursor
+#  Cursor is an AI-powered code editor. `.cursorignore` specifies files/directories to
+#  exclude from AI features like autocomplete and code analysis. Recommended for sensitive data
+#  refer to https://docs.cursor.com/context/ignore-files
+.cursorignore
+.cursorindexingignore
+
+# Marimo
+marimo/_static/
+marimo/_lsp/
+__marimo__/
+
+# Version py created by hatch-vcs
+src/clawmeter/_version.py
+

clawmeter-0.7.4/CHANGELOG.md

@@ -0,0 +1,135 @@
+# Changelog
+
+All notable changes to clawmeter are documented in this file.
+
+Format follows [Keep a Changelog](https://keepachangelog.com/en/1.1.0/).
+Versioning follows [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
+
+## [0.7.1] - 2026-04-10
+
+### Added
+
+- **Claude extra usage spend** (alpha) — tracks dollar spend beyond the subscription cap via undocumented `extra_usage` field in `/api/oauth/usage` response
+- Extra Usage window with percentage utilisation, raw spend and limit in user's billing currency
+- `extras` dict includes `extra_usage_enabled`, `extra_usage_spent`, `extra_usage_limit`
+- **Weekly Sonnet (7d)** window — new per-model utilisation window from Claude API (`seven_day_sonnet`)
+- **`credits` unit type** in formatters — displays as `$X.XX` with no currency qualifier, for values in the user's billing currency
+- Alpha warning utility (`emit_alpha_warning`) extracted to `config.py` as shared infrastructure for all alpha providers
+
+### Changed
+
+- Claude test fixture updated with `seven_day_sonnet`, `seven_day_cowork`, and `extra_usage` fields matching current API response
+- OQ-001, OQ-002, OQ-004, OQ-008 closed with research findings in SPEC.md
+
+## [0.7.0] - 2026-04-10
+
+### Added
+
+- **Ollama provider** (`providers/ollama.py`) — local instance monitoring with model inventory, loaded model state, VRAM/RAM usage reporting
+- Multi-host support: single `host` config or `[[providers.ollama.hosts]]` array for monitoring multiple Ollama instances across a network
+- Per-host polling via `GET /api/tags` (model inventory + health) and `GET /api/ps` (loaded models + memory allocation)
+- Cloud model detection via `:cloud` tag suffix in model names (labelling only)
+- Error isolation per host — one host being unreachable does not affect others
+- **Alpha features framework** (D-053) — `enable_alpha_features` config flag gates unstable data sources with stderr warning, `alpha: true` labelling in extras, and graceful failure
+- Ollama Cloud session/weekly usage windows (alpha) — probes `ollama.com/api/account/usage` when `cloud_enabled = true` and alpha flag set
+- Cloud API key authentication via credential chain (`api_key_command` > `api_key_env`/`$OLLAMA_API_KEY` > keyring)
+- `is_alpha_enabled()` config helper for providers to check alpha flag
+- `[providers.ollama]` config section with `host`/`hosts` mutual exclusivity validation
+- Research report: `docs/research/ollama-v0.7.0-research.md`
+
+### Fixed
+
+- Table and monitor formatters now render count and MB values (e.g. "3", "5,086 MB") instead of showing "0%" for non-percentage usage windows
+
+## [0.6.0] - 2026-04-10
+
+### Added
+
+- **OpenAI provider** (`providers/openai.py`) — organisation-level spend and per-model usage monitoring via the Administration API
+- Usage API: `GET /v1/organization/usage/completions` with `group_by=model` for per-model token counts (input, output, cached, requests)
+- Costs API: `GET /v1/organization/costs` with `group_by=line_item` for per-model cost in USD and total MTD spend
+- Admin key credential resolution via `$OPENAI_ADMIN_KEY` (`sk-admin-*`) with 4-tier chain
+- Token and cost merge into unified `ModelUsage` entries with `top_model_spend` extras
+- Config section for OpenAI (`providers.openai` with `admin_key_env`)
+
+### Changed
+
+- SPEC.md Section 3.3 rewritten with correct admin key auth, response schemas, and query parameters
+- OQ-012 closed: undocumented billing endpoints confirmed dead, admin key required
+- A-006 falsified: standard API keys cannot access org usage endpoints
+
+## [0.5.0] - 2026-04-09
+
+### Added
+
+- **xAI Grok provider** (`providers/grok.py`) — spend monitoring via the xAI Management API
+- Invoice preview endpoint for month-to-date spend with per-model cost breakdown
+- Spending limits endpoint with utilisation percentage and threshold detection
+- Prepaid balance endpoint
+- Usage analytics endpoint for per-model time-series token and spend data
+- Dual credential support: management key (primary, 4-tier resolution) and optional API key
+- Team ID resolution from config (`providers.grok.team_id`) or `$XAI_TEAM_ID` env var
+- Redaction pattern for `xai-*` keys in security module
+- Config section for Grok provider
+- Test fixtures for all four Management API responses
+
+### Fixed
+
+- Table and monitor formatters now render USD-based usage windows as dollar values instead of percentages
+
+## [0.4.0] - 2026-04-08
+
+### Added
+
+- **Monitor TUI** (`--monitor` flag) — Rich Live dashboard with auto-refreshing provider status
+- Per-provider status panels with usage bars, model breakdowns, and sparklines
+- Compact mode for smaller terminals
+- Configurable refresh interval
+- Colour-coded threshold indicators (normal/warning/critical)
+
+## [0.3.0] - 2026-04-07
+
+### Added
+
+- **Background daemon** (`clawmeter daemon start/stop/status/install`) — polls providers on a schedule and writes to the shared SQLite history database
+- PID file management with stale PID detection
+- `daemon install` generates a systemd user unit for automatic startup
+- CLI reads from daemon's database when available, falls back to direct fetching
+- **Docker deployment** — `Dockerfile` and `docker-compose.yml` for containerised daemon operation
+- Container mode detection (skips permission checks, uses env var paths)
+- Per-provider poll interval override in config
+- Daemon-aware provider base class with backoff state persistence
+
+## [0.2.0] - 2026-04-07
+
+### Added
+
+- **SQLite history store** (`history.py`) — records provider snapshots with WAL mode and retention pruning
+- `clawmeter history` — query historical usage data with time range filters
+- `clawmeter report` — generate usage summaries (daily/weekly/monthly) with trend analysis
+- `clawmeter export` — export history to CSV or JSON
+- Meaningful-change detection to avoid storing duplicate snapshots
+- Per-model usage tracking in dedicated `model_usage` table
+
+## [0.1.0] - 2026-04-06
+
+### Added
+
+- **Anthropic Claude provider** — subscription utilisation tracking via Claude Code OAuth credentials
+- Read-only consumer of `~/.claude/.credentials.json` (never writes to the file)
+- Five usage windows: Session (5h), Daily, Weekly, Monthly, and Opus-specific
+- Per-model usage breakdown (Opus, Sonnet, Haiku token counts)
+- **CLI with three output modes**: JSON (default, no flag), `--now` (Rich table), `--monitor` (TUI, added in v0.4.0)
+- **Provider architecture** — abstract `Provider` base class with `@register_provider` decorator
+- 4-tier credential resolution: `key_command` > env var > keyring > provider-specific file
+- `SecretStr` wrapper with safe `__repr__` — secrets never leak in logs or output
+- **JSON cache** with `poll_interval`-based TTL and `fcntl.flock()` file locking
+- **TOML configuration** from `~/.config/clawmeter/config.toml` with env var path overrides
+- **Security module** — credential sanitisation, secure file I/O (`0o600`/`0o700`), atomic writes
+- Allowed-hosts enforcement for credential-bearing HTTP requests
+- `key_command` execution with `shell=False` and 10s timeout
+- Rich table formatter with TTY-adaptive column widths and colour-coded thresholds
+- JSON formatter with stable schema (stdout-only, stderr for messaging)
+- Exit codes: 0 (ok), 1 (config error), 2 (all auth fail), 3 (partial), 4 (all network fail)
+- CI pipeline via GitHub Actions (lint, test, type check)
+- Full test suite: providers, formatters, cache, config, security, CLI, models

clawmeter-0.7.4/CLAUDE.md

@@ -0,0 +1,101 @@
+# CLAUDE.md
+
+This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
+
+## Project Overview
+
+clawmeter is a Linux-native CLI tool for monitoring LLM service usage, costs, and performance across multiple providers. It uses a pluggable provider architecture, launching with Anthropic Claude support and expanding to Grok (xAI), OpenAI, Ollama, and local system metrics. A GTK/GNOME desktop widget is planned for v2.
+
+The tool operates in two modes: **standalone** (CLI fetches directly) and **daemon** (background service collects, CLI reads from DB). The daemon is recommended for continuous history collection.
+
+The full specification lives in `docs/SPEC.md` - consult it for detailed provider specs, security requirements, CLI interface design, and open questions.
+
+## Build & Development
+
+```bash
+# Setup (from source)
+uv sync
+
+# Run
+uv run clawmeter
+
+# Run tests
+uv run pytest
+uv run pytest tests/test_security.py           # single test file
+uv run pytest tests/test_security.py::test_name # single test
+
+# Build
+uv build
+
+# Docker
+docker build -t clawmeter .
+docker compose up -d
+```
+
+Build backend is `hatchling` + `hatch-vcs`. Version is derived from git tags (no VERSION file). Tag format: `v0.1.0`.
+
+## Architecture
+
+**Daemon + thin clients:** The daemon polls providers on a schedule, writes to a SQLite history database. CLI/TUI/GTK are thin readers of that database. Without the daemon, the CLI falls back to direct fetching (standalone mode).
+
+**Provider abstraction** is the core pattern: every LLM service implements `Provider` (ABC) with `name()`, `display_name()`, `is_configured()`, `fetch_usage()`, and `auth_instructions()`. The base class handles credential resolution. Providers return a unified `ProviderStatus` dataclass containing `UsageWindow` and `ModelUsage` entries.
+
+Key components and their locations under `src/clawmeter/`:
+
+- `daemon.py` - Background service: poll loop, PID file management, systemd install
+- `models.py` - `SecretStr`, `UsageWindow`, `ModelUsage`, `ProviderStatus` dataclasses
+- `providers/base.py` - Abstract `Provider` class with `resolve_credential()` (4-tier: key_command > env var > keyring > provider-specific file). key_command failure raises `CredentialError` (no silent fallthrough). Provider registry via `@register_provider` decorator and module-level dict.
+- `providers/claude.py` - Read-only consumer of `~/.claude/.credentials.json` (`claudeAiOauth.accessToken` + `expiresAt`). Does NOT refresh tokens or write to the file.
+- `core.py` - Orchestrator: loads providers, aggregates results
+- `cli.py` - Click/Typer CLI with modes: JSON (default, no flag), `--now` (table), `--monitor` (Rich Live TUI)
+- `security.py` - Credential sanitisation, secure file I/O (`secure_write`, `secure_mkdir`), permission warnings
+- `cache.py` - Per-provider JSON cache (standalone mode only), `poll_interval`-based TTL, `fcntl.flock()`
+- `config.py` - TOML loader from `~/.config/clawmeter/config.toml`, permission warnings (not hard failures)
+- `history.py` - SQLite at `~/.local/share/clawmeter/history.db` (WAL mode), retention pruning, reporting
+- `formatters/` - JSON, Rich table (TTY-adaptive), Rich Live TUI
+
+## Critical Design Decisions
+
+- **stdout is data only, stderr is messaging only** - JSON/table output to stdout; errors, warnings, spinners to stderr. No exceptions.
+- **JSON is the default output mode** (no flag needed). `--now` for table, `--monitor` for TUI.
+- **Daemon decouples collection from presentation** - CLI/TUI/GTK read from the shared SQLite DB. Standalone mode still works without daemon.
+- **No plaintext secrets in config files** - credentials resolved via keyring, env vars, or `key_command` (executed with `shell=False`).
+- **All secrets wrapped in `SecretStr`** - `__repr__` always returns `SecretStr('***')` (never leaks real characters). Never log/serialize raw secrets. Failures raise `CredentialError` (defined in models.py).
+- **Read-only consumer of Claude credentials** - the tool NEVER writes to `~/.claude/.credentials.json`. On token expiry, direct user to `claude /login`.
+- **key_command failure is hard, not silent** - if explicitly configured and it fails/times out, raise `CredentialError`. Don't silently fall through to env vars.
+- **Global poll_interval (10m default)** - one setting replaces the old `cache_ttl`/`refresh_interval` split. Per-provider override available. Local providers default to 60s.
+- **Async concurrency** - `Provider.fetch_usage()` is async. CLI uses `asyncio.run()` + `asyncio.gather()` for concurrent provider fetches. Daemon uses a persistent asyncio event loop. `return_exceptions=True` for error isolation.
+- **Rate-limit backoff** - exponential backoff on 429s (10m -> 20m -> 40m, cap 60m), persisted in cache file, resets on success.
+- **Files created with `0o600`/dirs `0o700` from the start** - use `os.open()` + `os.fdopen()`, not `open()` then `chmod()`.
+- **Atomic file writes** - write to `.tmp` then `os.rename()`.
+- **Permission checks are warnings, not hard failures** - config contains no secrets (by design). Container mode skips checks entirely.
+- **Provider errors are isolated** - one provider failing doesn't block others. Exit codes: 0=all ok, 1=config error, 2=all auth fail, 3=partial, 4=all network fail.
+- **Australian English in docs/UI** ("utilisation", "colour") but US English for Python API identifiers where convention requires it.
+- **Lightweight base install, additive extras** - base has no C extensions. `[local]` adds psutil/pynvml. `[gtk]` adds PyGObject. `[all]` is everything.
+- **Env var overrides for all paths** - `CLAWMETER_CONFIG`, `CLAWMETER_DATA_DIR`, `CLAWMETER_CACHE_DIR`. Essential for Docker.
+
+## Security Checklist
+
+When touching credential or network code, verify:
+- `SecretStr` wraps all secrets immediately on read — `__repr__` is always `SecretStr('***')`
+- Redaction patterns applied to any user-visible output (see SPEC Section 7.3)
+- `httpx` requests: `verify=True` (TLS), `follow_redirects=False` on auth-bearing requests
+- `key_command`: `shell=False`, timeout=10s with `TimeoutExpired` handling, only log stderr on failure (never stdout)
+- Credential-bearing requests only sent to provider's `allowed_hosts`
+- No secrets in cache files, JSON output, `extras` dict, or error messages
+- Never write to Claude's credential file — read-only consumer
+
+## Tech Stack
+
+Python 3.10+, httpx, rich, click/typer, keyring, tomllib/tomli, sqlite3, hatchling+hatch-vcs, uv, pytest+respx, Docker, systemd user units. Optional: psutil, pynvml (via `[local]` extra), PyGObject (via `[gtk]` extra).
+
+## Release Milestones
+
+v0.1.0: CLI MVP — Claude provider, standalone, JSON + table output, CI from day one
+v0.2.0: History + reporting — SQLite store, meaningful-change detection, report/export commands
+v0.3.0: Daemon + Docker — background collection, systemd, container deployment
+v0.4.0: Monitor TUI — Rich Live dashboard, sparklines, compact mode
+v0.5.0–v0.8.0: Providers — Grok, OpenAI, Ollama, Local system metrics
+v0.9.0: Notifications and polish
+v1.0.0: Stable release — security audit, schema freeze, fuzz testing
+v2.0.0: GTK4/libadwaita desktop widget

clawmeter-0.7.4/Dockerfile

@@ -0,0 +1,24 @@
+FROM python:3.12-slim AS builder
+
+RUN apt-get update && apt-get install -y --no-install-recommends git && rm -rf /var/lib/apt/lists/*
+COPY . /tmp/build
+RUN pip install --no-cache-dir build && python -m build --wheel /tmp/build -o /tmp/dist
+
+FROM python:3.12-slim
+
+# Create non-root user
+RUN useradd --create-home --shell /bin/bash monitor
+
+COPY --from=builder /tmp/dist/*.whl /tmp/
+RUN pip install --no-cache-dir /tmp/*.whl && rm -f /tmp/*.whl
+
+USER monitor
+
+# Default data directory
+ENV CLAWMETER_DATA_DIR=/data
+ENV CLAWMETER_CACHE_DIR=/data/cache
+ENV CLAWMETER_CONTAINER=1
+
+VOLUME /data
+
+ENTRYPOINT ["clawmeter", "daemon", "run"]

clawmeter-0.7.4/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 Daniel Thomas
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.