codex-usage-tracking 0.21.0__tar.gz → 0.23.0__tar.gz

codex_usage_tracking-0.23.0/.codex-plugin/plugin.json

@@ -0,0 +1,45 @@
+{
+  "name": "codex-usage-tracker",
+  "version": "0.23.0",
+  "description": "Unofficial local, evidence-backed Codex usage analyst with MCP tools and an Evidence Console.",
+  "author": {
+    "name": "Douglas Monsky"
+  },
+  "homepage": "https://github.com/douglasmonsky/codex-usage-tracker",
+  "repository": "https://github.com/douglasmonsky/codex-usage-tracker",
+  "license": "MIT",
+  "keywords": [
+    "codex",
+    "tokens",
+    "usage",
+    "mcp",
+    "dashboard"
+  ],
+  "skills": "./skills/",
+  "mcpServers": "./.mcp.json",
+  "interface": {
+    "displayName": "Codex Usage Tracker",
+    "shortDescription": "Local, evidence-backed Codex usage analyst with MCP tools and an Evidence Console",
+    "longDescription": "Unofficial independent project, not made by, affiliated with, endorsed by, sponsored by, or supported by OpenAI. Reads local Codex session logs to power MCP-based usage analysis, allowance diagnostics, token-waste investigations, shareable aggregate reports, and a supporting local Evidence Console without uploading logs.",
+    "developerName": "Douglas Monsky",
+    "category": "Productivity",
+    "capabilities": [
+      "Interactive",
+      "Read",
+      "Write"
+    ],
+    "websiteURL": "https://github.com/douglasmonsky/codex-usage-tracker",
+    "privacyPolicyURL": "https://github.com/douglasmonsky/codex-usage-tracker",
+    "termsOfServiceURL": "https://github.com/douglasmonsky/codex-usage-tracker",
+    "defaultPrompt": [
+      "What drove my Codex usage this week?",
+      "Look through my usage for token waste",
+      "Compare model and effort usage",
+      "Check whether my weekly allowance changed"
+    ],
+    "brandColor": "#2563EB",
+    "composerIcon": "./assets/icon.svg",
+    "logo": "./assets/icon.svg",
+    "screenshots": []
+  }
+}

codex_usage_tracking-0.23.0/.mcp.json

@@ -0,0 +1,13 @@
+{
+  "mcpServers": {
+    "codex-usage-tracker": {
+      "command": "python3",
+      "args": ["./skills/codex-usage-tracker/scripts/run_mcp.py"],
+      "cwd": ".",
+      "env": {
+        "CODEX_USAGE_TRACKER_MCP_PROFILE": "core"
+      },
+      "startup_timeout_sec": 120
+    }
+  }
+}

codex_usage_tracking-0.23.0/AGENTS.md

@@ -0,0 +1,294 @@
+# Codex Usage Tracker Instructions
+
+## Project Purpose
+
+This repo builds a local Codex plugin and dashboard that track aggregate token usage from Codex session logs.
+
+## Tech Stack
+
+- Python 3.10+
+- SQLite via the Python standard library
+- MCP Python SDK for Codex tool exposure
+- Pytest for tests
+
+## Repo Layout
+
+- `src/codex_usage_tracker/` - parser, SQLite store, reports, dashboard, CLI, and MCP server.
+- `src/codex_usage_tracker/context.py` - on-demand raw-context reader for one selected usage record.
+- `src/codex_usage_tracker/reports.py` - shared application/report services used by CLI and MCP wrappers.
+- `src/codex_usage_tracker/api_payloads.py` - shared stable JSON payload builders for CLI and MCP surfaces.
+- `src/codex_usage_tracker/schema.py` - single source of truth for persisted usage-event columns.
+- `src/codex_usage_tracker/threads.py` - thread attachment inference used by dashboard payload generation.
+- `src/codex_usage_tracker/pricing_config.py`, `pricing_openai.py`, `pricing_estimates.py`, and `costing.py` - pricing config, source parsing, estimate policy, and cost calculations behind the `pricing.py` facade.
+- `src/codex_usage_tracker/allowance.py` - Codex credit-rate and optional local allowance-window helpers.
+- `src/codex_usage_tracker/plugin_installer.py` - package-owned local Codex plugin installer.
+- `src/codex_usage_tracker/plugin_data/` - plugin assets, dashboard template/assets, local dashboard guide, screenshots, and skill files bundled into wheels.
+- `skills/codex-usage-tracker/` and `src/codex_usage_tracker/plugin_data/skills/codex-usage-tracker/` - operational Codex skill for tracker setup, summaries, dashboard generation, and MCP tools.
+- `skills/codex-usage-api/` and `src/codex_usage_tracker/plugin_data/skills/codex-usage-api/` - companion Codex skill for conversational analysis using the stable JSON API/MCP tools.
+- `src/codex_usage_tracker/server.py` - localhost dashboard server with live aggregate refresh and lazy context endpoints.
+- `~/.codex-usage-tracker/pricing.json` - optional local-only pricing config, never committed.
+- `~/.codex-usage-tracker/allowance.json` - optional local-only copied allowance state, never committed.
+- `.codex-plugin/plugin.json` - Codex plugin manifest.
+- `.mcp.json` - MCP server configuration for Codex.
+- `scripts/install_local_plugin.py` - compatibility wrapper around `codex-usage-tracker install-plugin`.
+- `scripts/check_release.py` - release-readiness checks for docs, versions, packaging, wheel contents, and tracked secret patterns.
+- `.github/workflows/ci.yml` - GitHub Actions test and package build workflow.
+- `.github/workflows/pricing-compat.yml` - scheduled/manual non-blocking live pricing parser compatibility check.
+- `docs/` - install, dashboard, CLI, pricing/credits, MCP, privacy, architecture, development, JSON-schema docs, and screenshots built from synthetic aggregate fixture data.
+- `tests/` - synthetic fixtures and unit tests.
+
+## Setup
+
+```bash
+python3 -m venv .venv
+. .venv/bin/activate
+python -m pip install --upgrade pip
+python -m pip install ".[dev]" twine
+codex-usage-tracker install-plugin --python .venv/bin/python
+```
+
+## Branch And PR Workflow
+
+This project is now a published PyPI package with user-facing docs, JSON/MCP contracts, a release workflow, and privacy guarantees. Treat `main` as always releasable.
+
+### MCP-First Pivot Execution
+
+Work in the MCP-first pivot must follow
+`docs/roadmap/mcp-first-pivot.md` and its approved detailed implementation
+roadmap. Use one focused `pivot/<task-number>-<slug>` branch per task, implement
+only the task's declared interfaces, and update
+`docs/roadmap/mcp-first-pivot-execution.md` in the same commit with branch,
+commits, focused and full verification, deviations, and follow-up risks.
+
+Do not add a dashboard workspace, top-level MCP concept, top-level CLI command,
+runtime dependency, or SQLite table unless the roadmap names it or an approved
+design amendment authorizes it. Compatibility removal must also be due in
+`docs/deprecations.md`.
+
+- Do not commit directly to `main`.
+- Start each coherent task from current `main` with a short-lived branch.
+- Use branch prefixes `feature/`, `fix/`, `docs/`, `chore/`, `test/`, `release/`, `hotfix/`, or `pivot/`. Reserve `pivot/` for tasks in the approved MCP-first roadmap.
+- Keep each branch focused on one issue, one reviewable task, or one release.
+- Do not create a long-lived `develop` branch.
+- Do not mix release prep with unrelated feature work.
+- Push task branches and open a PR for all changes headed to `main`.
+- Prefer squash merge for ordinary task PRs so `main` stays readable.
+- Use the PR as the review artifact even when there is only one maintainer.
+
+Recommended branch names:
+
+```text
+feature/<issue-number>-short-description
+fix/<issue-number>-short-description
+docs/<issue-number>-short-description
+chore/<issue-number>-short-description
+test/<issue-number>-short-description
+release/0.4.0
+hotfix/0.3.3
+```
+
+Before starting a task branch:
+
+```bash
+git switch main
+git pull --ff-only
+git switch -c docs/123-short-description
+```
+
+## Agent Boundaries
+
+Codex may create task branches, write tests, update docs, run local gates, prepare PR summaries, prepare release branches, and prepare changelog/version changes.
+
+Codex must not do these without explicit maintainer approval:
+
+- Push directly to `main`.
+- Create or push release tags.
+- Publish to TestPyPI or PyPI.
+- Add PyPI or TestPyPI API tokens.
+- Publish from a local machine.
+- Change privacy semantics.
+- Rename the PyPI distribution, import package, CLI command, plugin name, MCP tools, schema IDs, or stable JSON contracts.
+- Delete branches.
+- Force-push shared branches.
+
+Publishing must happen only through the approved GitHub Actions Trusted Publishing workflow and protected `testpypi`/`pypi` environments.
+
+## Issue And Milestone Workflow
+
+Use GitHub issues as the normal unit of work once the task is non-trivial. A branch should usually map to one issue and close it from the PR.
+
+Recommended labels:
+
+```text
+bug
+docs
+packaging
+release
+privacy
+security
+performance
+dashboard
+cli
+mcp
+parser-compat
+good-first-issue
+blocked
+1.0-blocker
+```
+
+Recommended milestones:
+
+```text
+0.4.0
+1.0-readiness
+1.0.0
+```
+
+Use patch releases for public blockers such as broken PyPI installs, missing package data, broken CLI entry points, privacy leaks, bad plugin installs, or bad runtime pins. Put planned stabilization work into the next minor release instead of bundling it into a patch.
+
+## Performance Work
+
+Use the `agent-perf` skill and CLI when optimizing localhost API endpoints, dashboard refreshes, report generation, content indexing, parser or SQLite hot paths, or other latency- or CPU-sensitive code, and when investigating a performance regression or making a speedup claim.
+
+Install the repository's pinned profiling tools with `uv sync --group performance`. Profile the smallest repeatable workload with synthetic or anonymized inputs; never profile production, live private databases, arbitrary processes, or real Codex session content. Record the identical workload without a profiler before making a performance claim. Change one suspected cause at a time, rerun the unprofiled workload, and use `agent-perf compare` only to compare attribution evidence rather than as proof of a speedup.
+
+## Validation
+
+Run focused tests first, then broader checks. Run the full local CI gate before opening or updating PRs that touch release, packaging, CLI contracts, MCP behavior, dashboard behavior, privacy behavior, schemas, generated docs/assets, or bundled plugin/skill files.
+
+## Source Inspection And Tool Output
+
+Large command outputs in Codex chat can be visually compacted by the transcript renderer. When inspecting source, especially after broad `rg`, `sed`, `nl`, generated dashboard assets, logs, or workflow output, do not treat a mangled rendered snippet as proof that the file is corrupt. Prefer small targeted file windows, `git diff`, `python -m py_compile`, focused tests, and CI as the source of truth. If exact syntax matters, inspect a narrow range or use a parser/compiler rather than relying on large printed source dumps.
+
+```bash
+python -m ruff check .
+python -m mypy
+python -m pytest
+python -m pytest --cov=codex_usage_tracker --cov-report=term-missing
+python -m compileall src
+for file in src/codex_usage_tracker/plugin_data/dashboard/dashboard*.js; do
+  node --check "$file"
+done
+python scripts/check_release.py
+git diff --check
+rm -rf dist build src/codex_usage_tracker.egg-info src/codex_usage_tracking.egg-info
+python -m build
+python -m twine check dist/*
+python scripts/check_release.py --dist
+```
+
+Additional smoke checks for touched CLI surfaces:
+
+```bash
+python -m pytest
+python -m compileall src
+for file in src/codex_usage_tracker/plugin_data/dashboard/dashboard*.js; do
+  node --check "$file"
+done
+python -m build
+python scripts/check_release.py --dist
+git diff --check
+python scripts/smoke_installed_package.py
+codex-usage-tracker update-pricing --output /tmp/codex-usage-pricing.json
+codex-usage-tracker update-rate-card --output /tmp/codex-usage-rate-card.json
+codex-usage-tracker doctor
+codex-usage-tracker doctor --suggest-repair
+codex-usage-tracker dashboard --output /tmp/codex-usage-dashboard.html
+codex-usage-tracker serve-dashboard --help
+codex-usage-tracker init-allowance --output /tmp/codex-usage-allowance.json
+codex-usage-tracker parse-allowance --output /tmp/codex-usage-allowance.json "5h 79% 6:50 PM Weekly 33% Jun 7"
+codex-usage-tracker init-thresholds --output /tmp/codex-usage-thresholds.json
+codex-usage-tracker init-projects --output /tmp/codex-usage-projects.json
+codex-usage-tracker support-bundle --output /tmp/codex-usage-support.json
+codex-usage-tracker pricing-coverage
+codex-usage-tracker summary --preset by-subagent-role
+codex-usage-tracker expensive --limit 5
+```
+
+For documentation-only branches, at minimum run:
+
+```bash
+python scripts/check_release.py
+git diff --check
+```
+
+## Release Branches
+
+Use release branches only for version/changelog/pinning/publish prep, for example `release/0.4.0` or `hotfix/0.3.3`.
+
+Release branches may include:
+
+- Version bumps.
+- `CHANGELOG.md` updates.
+- Install/version wording updates.
+- Runtime package pins.
+- Publish workflow tweaks.
+- Release notes.
+- Final smoke-test fixes directly tied to release readiness.
+
+Release branches must not include unrelated features.
+
+Recommended release sequence:
+
+```bash
+git switch main
+git pull --ff-only
+git switch -c release/0.4.0
+# version/changelog/release edits
+python -m ruff check .
+python -m mypy
+python -m pytest
+python -m pytest --cov=codex_usage_tracker --cov-report=term-missing
+python -m compileall src
+for file in src/codex_usage_tracker/plugin_data/dashboard/dashboard*.js; do
+  node --check "$file"
+done
+python scripts/check_release.py
+git diff --check
+rm -rf dist build src/codex_usage_tracker.egg-info src/codex_usage_tracking.egg-info
+python -m build
+python -m twine check dist/*
+python scripts/check_release.py --dist
+git add .
+git commit -m "Prepare 0.4.0 release"
+git push -u origin release/0.4.0
+```
+
+Open a PR to `main` and merge only after CI passes. After merge, tag from updated `main`, not from an unreviewed release branch, and only after explicit maintainer approval:
+
+```bash
+git switch main
+git pull --ff-only
+git tag -a v0.4.0 -m "codex-usage-tracker 0.4.0"
+git push origin v0.4.0
+```
+
+## Privacy Rules
+
+- Never commit real Codex session logs.
+- Never commit real prompts, assistant text, tool outputs, pasted secrets, message snippets, or raw Codex logs.
+- The local content index may store bounded snippets in the user-owned SQLite database by approved design. Do not expose indexed/raw content through default CSV, generated HTML, support bundles, screenshots, aggregate JSON, fixtures based on real logs, or commits.
+- Raw context may be read during refresh/content indexing or explicit selected-call context loading. Keep shareable outputs aggregate-first unless a command is explicitly documented as a local raw/content export.
+- Keep fixture data synthetic.
+- Keep local SQLite databases, CSV exports, HTML dashboards, caches, and virtualenvs out of git.
+- Do not hard-code real current USD model pricing in source; refresh the local config from OpenAI's published pricing docs or use manual local overrides. Internal Codex model estimates must be explicitly marked as estimates with source and rationale metadata.
+- Source-stamped Codex credit rate-card snapshots must include source/date metadata, confidence labels, and local override support. Manually copied allowance remaining values stay in local config only.
+
+## Definition Of Done
+
+- Parser and content-index handling are covered by synthetic session logs.
+- SQLite refresh is idempotent.
+- MCP tool functions return concise aggregate data by default; content-aware tools must be explicit local investigation surfaces.
+- Dashboard generated HTML is aggregate-first and does not embed indexed/raw content.
+- Doctor, summary presets, dashboard, and expensive-call views work from CLI and MCP wrappers.
+- `codex-usage-tracker install-plugin` can register the installed package without relying on a source-checkout symlink.
+- `python -m codex_usage_tracker` and `codex-usage-tracker --version` both work.
+- Wheel and source distribution builds include plugin assets and the Codex skill.
+- `scripts/check_release.py --dist` passes before any public release.
+- Pricing coverage clearly separates configured, estimated, and unpriced model usage.
+- Codex credit coverage clearly separates exact rate-card matches, inferred aliases, and missing credit rates.
+- Dashboard Calls and Threads views share filters, totals, and aggregate-first hover details.
+- Dashboard usage docs are updated when the visible dashboard workflow changes, and screenshots must be generated from synthetic data only.
+- Dashboard refresh is localhost-only, generated HTML stays aggregate-first, and context loading is lazy, localhost-only, explicit, redacted, and not embedded in the static HTML payload.
+- Subagent calls preserve logged parent-session metadata, latch to parent thread labels when available, and auto-review attachment is clearly marked when inferred.
+- Tests and compile checks pass.