Show HN: Decant – Understand how you spend tokens

Dosu released Decant, a local-first CLI and web tool that analyzes Claude Code and Codex session logs to show token usage, estimated cost, context, and agent activity, with full-text search and exports. The tool runs via npx, npm, Homebrew, or Docker, supports macOS and Linux on x64 and arm64, and makes no outbound network calls at runtime.

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. 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. Run Decant with npx with no Bun install or global package required. npx @dosu/decant@latest This starts the local UI at http://127.0.0.1:3000 http://127.0.0.1:3000 , indexes the Claude Code and Codex logs on your machine, and watches for changes. - One SQLite archive for Claude Code and Codex sessions. - Full-text search across messages, tool calls, and transcripts. - Token, estimated cost /dosu-ai/decant/blob/main/docs/pricing.md , context, activity, tool, MCP, and file analytics. - Browsable sessions, projects, files, and ingest diagnostics. - Markdown, JSON, report, and trajectory exports. - Deterministic scripts, replays, and agent instructions distilled from command history. Published binaries support macOS and Linux on x64 and arm64. Native Windows binaries are not currently available. Install a persistent command with npm: npm install --global @dosu/decant@latest decant Or use Homebrew: brew install dosu-ai/dosu/decant decant Or install the latest release without Node.js or Bun: curl -fsSL https://raw.githubusercontent.com/dosu-ai/decant/main/install.sh | sh decant See Distribution /dosu-ai/decant/blob/main/docs/distribution.md for installer options, Docker, source builds, and release verification. docker run --rm \ -p 127.0.0.1:3000:3000 \ -v decant-data:/var/lib/decant \ -v "$HOME/.claude/projects:/sources/claude:ro" \ -v "$HOME/.codex:/sources/codex:ro" \ ghcr.io/dosu-ai/decant:latest Keep the 127.0.0.1: prefix. Publishing -p 3000:3000 exposes the unauthenticated archive API on every host interface. Custom container networks may need the trusted-peer setting documented under Docker distribution /dosu-ai/decant/blob/main/docs/distribution.md docker . decant start the local web UI decant sync index new and changed sessions decant ls list sessions decant show 1 render a transcript decant search "auth bug" full-text search decant stats --by model usage and cost rollups decant economics token, cost, and time breakdowns decant files --group ext file hotspots decant tool stats tool usage decant mcp stats MCP server usage decant export 1 session.md export a session Run decant --help or decant <command --help for all commands and flags. Read commands support --json ; global flags include --db , --format , --quiet , --no-color , and --no-sync . To index selected files or a temporary source tree: decant --db /tmp/decant.db sync --path ./session.jsonl --path ./sessions The OpenAPI 3.1 contract is docs/api/openapi.yaml /dosu-ai/decant/blob/main/docs/api/openapi.yaml , and a running server exposes it at http://127.0.0.1:3000/api/openapi.json http://127.0.0.1:3000/api/openapi.json . See API recipes /dosu-ai/decant/blob/main/docs/api/recipes.md for examples. Analytics methodology /dosu-ai/decant/blob/main/docs/analytics-methodology.md Pricing estimates /dosu-ai/decant/blob/main/docs/pricing.md Archive and data lifecycle /dosu-ai/decant/blob/main/docs/data-lifecycle.md Local Serve API /dosu-ai/decant/blob/main/docs/api/routes.md Distribution and release verification /dosu-ai/decant/blob/main/docs/distribution.md Architecture /dosu-ai/decant/blob/main/docs/architecture.md Contributions are welcome. Read CONTRIBUTING.md /dosu-ai/decant/blob/main/CONTRIBUTING.md for setup, tests, and privacy requirements. Coding agents should also read AGENTS.md /dosu-ai/decant/blob/main/AGENTS.md . Use synthetic session data in issues and tests. Real transcripts can contain source code, prompts, credentials, and local paths. Report vulnerabilities privately as described in SECURITY.md /dosu-ai/decant/blob/main/SECURITY.md . Decant is built and maintained by Dosu https://dosu.dev?utm source=decant&utm medium=github&utm campaign=attribution&utm content=about , Knowledge Infrastructure for Agents. Dosu helps make agents faster, cheaper, and more effective. Decant shows what your agents spent, read, and touched via your agent logs. Decant was inspired in part by Letta's Trajectory https://github.com/letta-ai/trajectory , which normalizes agent transcripts across runtimes into a shared record format. Decant is licensed under the Apache License 2.0 /dosu-ai/decant/blob/main/LICENSE .