Local-first MCP tools and dashboard for investigating Codex token usage, credits, costs, caching, and thread patterns.
# Add to your Claude Code skills
git clone https://github.com/douglasmonsky/codex-usage-trackerGuides for using mcp servers skills like codex-usage-tracker.
Last scanned: 7/15/2026
{
"issues": [],
"status": "PASSED",
"scannedAt": "2026-07-15T06:13:16.916Z",
"npmAuditRan": true,
"pipAuditRan": true,
"promptInjectionRan": true
}codex-usage-tracker is an open-source mcp servers skill for AI coding assistants such as Claude Code, Codex CLI, and ChatGPT, built by douglasmonsky. Local-first MCP tools and dashboard for investigating Codex token usage, credits, costs, caching, and thread patterns. It has 192 GitHub stars.
Yes. codex-usage-tracker passed SkillsLLM's automated security scan — a dependency vulnerability audit plus prompt-injection heuristics — with no high-severity issues. You can read the full report in the Security Report section on this page.
Clone the repository with "git clone https://github.com/douglasmonsky/codex-usage-tracker" and add it to your Claude Code skills directory (see the Installation section above).
codex-usage-tracker is primarily written in Python. It is open-source under douglasmonsky on GitHub, so you can review or fork the full source.
Yes. SkillsLLM lists many other MCP Servers skills you can browse and compare side by side. Open the MCP Servers category from the badge at the top of this page, or use the Related Skills and comparison links further down to weigh codex-usage-tracker against similar tools.
No comments yet. Be the first to share your thoughts!
Top skills in this category by stars
Codex Usage Tracker is a local usage-data product for answering questions about how Codex work consumes tokens, time, tools, allowance, credits, and cost. It turns local session metadata into exact, bounded facts and stable evidence that an agent can explore and explain.
Version 0.28.0 is the current public release. Its six-tool MCP surface, Evidence Console, and SQLite kernel remain supported while the replacement is qualified.
The next product is a clean, agent-first kernel rather than a larger dashboard:
The 0.28 code remains a frozen oracle during the rewrite and is removed before the clean replacement release. The new runtime does not migrate or open its database.
Start with the documentation authority, read the
product direction, or follow the
clean-cutover roadmap. Progress is
tracked in the checkbox ledger, with one
complete contract per file under docs/roadmap/tasks/.
Install the Python runtime into an isolated executable environment, then add the matching tagged repository as a Codex plugin marketplace:
pipx install "codex-usage-tracking==0.28.0"
codex plugin marketplace add douglasmonsky/codex-usage-tracker --ref v0.28.0
codex plugin add codex-usage-tracker@codex-usage-tracker
The plugin launches the codex-usage-tracker executable from the pipx
environment. Start a fresh Codex task after installation so it discovers the
new plugin tool catalog.
python3 -m venv .venv
.venv/bin/python -m pip install -e ".[dev]"
just v
Use synthetic fixtures only. Do not commit or share real Codex logs, prompts, responses, tool-output bodies, credentials, private paths, or local databases.