Decant — Development skill for Claude Code
Local-first analysis of Claude Code and Codex sessions: token spend, context windows, files touched, and cost.
How to install Decant
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open dosu-ai/decant and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Decant does
Local-first analysis of Claude Code and Codex sessions: token spend, context windows, files touched, and cost.
Alternatives in Development
- Update Scaffold — Check for and apply myPKA scaffold framework updates safely 296 ★
- Quosparr — 🐦 Native menu bar & tray app for macOS, Windows and Linux: track Codex, Claude Code and Grok quotas, resets 288 ★
- Edgee — Official Edgee CLI to interract with Edgee's Agent Gateway 127 ★
README
Decant
[](https://www.npmjs.com/package/@dosu/decant) [](https://www.npmjs.com/package/@dosu/decant) [](https://github.com/dosu-ai/decant/actions/workflows/ci.yml) [](https://scorecard.dev/viewer/?uri=github.com/dosu-ai/decant) [](LICENSE)
Decant turns the Claude Code and Codex session logs on your machine into tangible insights. See where tokens, cost, and agent time go, inspect context usage, find the files and tools agents touch, and search complete transcripts from a CLI or local web UI.
Decant is local-first. It makes no outbound network calls at runtime, and your transcripts never leave your machine.
It does keep a copy of them. To make sessions searchable, Decant writes prompts, tool inputs, tool output, and canonicalized raw records for retained transcript messages into an **unencrypted** SQLite archive at `~/.decant/decant.db`, with a full-text index over the prompt and tool-argument text. Whatever your agents read is in there too — source code, file contents, local paths, and any credentials pasted into a session. Decant creates the archive owner-only (`0600`). If your organization has a retention policy for agent transcripts, this archive is subject to it. See [What the archive stores](docs/data-lifecycle.md#what-the-archive-stores) to inspect, delete, or remove it.
Built by [Dosu](https://dosu.dev?utm_source=decant&utm_medium=github&utm_campaign=attribution&utm_content=readme), Knowledge Infrastructure for Agents. Dosu helps make agents faster, cheaper, and more effective.

Quick start
Run Decant
Related Skills
Flowmeter Public
Local, privacy-first token, cost, cache, and trace analysis for Cursor, Claude Code, and Codex. Optional Flow-
Agent Max
Live view of Claude Code agents and subagents in VS Code: the tree, per-agent context and token spend. Local o
Claude Token Weather
Claude Code mod: context-window forecast above the prompt with 5h/7d usage windows and session cost
Chatlens
Local-first CLI for searching Codex, Claude Code, and Hermes sessions and recovering work context.
Claude Boot Stats
Per-component token cost analyzer for Claude Code first-turn context: HTTP intercept proxy + Qwen3 BPE + groun
Claude Code Token Dashboard
Self-hosted dashboard for Claude Code token usage and API-equivalent cost. Reads your local transcripts, runs
Related Agents
Ade
Agent sessions routinely outlast the machine that started them, and checking a Claude Code or Codex run from a
Budget
AI cost engineering — LLM spend tracking, model cost optimization, budget alerts, token efficiency audits
Cost Governor
Track aggregate token spend and API costs across all agent operations. Enforce budget constraints, detect cost