weavr 1.1.0

Claude Code transcript exporter — beautiful, self-contained HTML and Markdown
Documentation

weavr

CI Crates.io Crates.io Downloads Rust License: MIT 100% local

A fast, self-contained Rust CLI that converts Claude Code transcript JSONL files into beautiful HTML and Markdown.

weavr is a Rust reimplementation of claude-code-log, focused on speed, zero-dependency output artefacts, and a clean Material 3 design.

weavr in action

Navigation

Why weavr?

claude-code-log proved how useful it is to turn raw Claude Code transcripts into readable artefacts. weavr is a ground-up Rust rewrite around two goals:

  • Speed. Single-pass JSONL parser, in-memory session DAG, and SQLite cache for incremental rebuilds. 18–46× faster than the Python tool (see Performance).
  • Design. Material 3 light/dark theme, flat dot-timeline, rich inline tool rendering (Bash IN/OUT, diffs, modals for Skill/Agent). Every HTML file is fully self-contained — fonts, CSS, JS all embedded. A CI gate rejects any external URL.

Single static binary via brew, cargo binstall, or cargo install, with built-in self-update. No Python runtime needed.

weavr vs. claude-code-log

weavr (Rust) claude-code-log (Python)
Speed (all projects) ~1.3 s ~24 s
Speed (single session) ~28 ms ~1.3 s
Distribution single static binary (brew / binstall / cargo) uvx / pip (needs Python)
Self-contained output ✅ zero external URLs (CI-enforced) partial
Incremental rebuilds ✅ SQLite cache ⚠️ re-renders each run
Theme Material 3 light/dark, dot-timeline minimalist HTML
HTML export ✅ ✅
Markdown export + detail levels + --compact ✅ ✅
Token usage tracking ✅ ✅
Date-range filtering (natural language) ✅ ✅
Client-side filter chips + in-page search ✅ ✅ (runtime filtering)
Multi-project hierarchy + master index ✅ ✅
Self-update command ✅ —

Both tools share the same JSONL input format. If you live in the terminal and want an interactive TUI today, claude-code-log is great. If you want the fastest possible export and offline-portable HTML, reach for weavr.

How it works

 ~/.claude/projects/
   ├── my-project/session.jsonl          (Claude Code transcripts)
   └── other-app/session.jsonl
        │
        ▼
   ┌─────────────────────────────────────────────┐
   │  weavr                                      │
   │  ─────────────────────────────────────────  │
   │  JSONL parser → session DAG → SQLite cache  │
   │                     │                       │
   │              HTML renderer                  │
   │        (Material 3 · dot-timeline ·         │
   │         tool cards · diffs · search)        │
   └─────────────────────────────────────────────┘
        │
        ▼
   weavr-out/
   ├── index.html                        (master index, searchable)
   ├── my-project/combined_transcripts.html
   └── my-project/session.html           (fully self-contained)

Every output file is a single portable HTML — fonts, CSS, JS all embedded. No server, no CDN, no runtime. No AI involved — your transcripts never leave your machine.

Quickstart

1. Install

brew install deepakkumardewani/weavr/weavr
# or: cargo binstall weavr
# or: cargo install weavr

2. Export a session

weavr -i ~/.claude/projects/my-project/session.jsonl
# → writes session.html

3. Export everything

weavr --all-projects --open-browser
# → walks ~/.claude/projects/, generates weavr-out/, opens index.html

Two modes

weavr has two distinct operating modes:

Mode Command Produces
Batch weavr / weavr --all-projects HTML index + per-session + combined pages
Single-file weavr export <INPUT> one file (HTML or Markdown)

Batch mode walks ~/.claude/projects/ and generates a complete static site. It's HTML-only — no --format / --detail / --compact flags here.

Single-file mode (weavr export) gives you full control over format, detail level, and compactness. All format-related flags live on the export subcommand:

weavr export --help   # see all export options

The -i shorthand is a convenience alias for export <INPUT> (HTML only, full detail).

Key Features

  • Self-Contained HTML — fonts, CSS tokens, assets all embedded inline; CI-enforced
  • Light/Dark Themes — Material 3 warm-neutral tokens; dark by default, toggle persisted to localStorage
  • Markdown Export — --format md with five --detail levels and --compact mode
  • Flat Dot-Timeline — chronological event stream: user messages, assistant text, thinking, tool calls
  • Rich Tool Rendering — Bash IN/OUT, Read/Edit/Write diffs with intra-line highlights, Skill/Agent modals
  • Token Usage Tracking — per-message and per-session input/output/cache tokens
  • Multi-Project Export — --all-projects generates a navigable static site with master index
  • SQLite Cache — fast incremental rebuilds; --clear-cache / --no-cache for control
  • Interactive HTML — filter chips, in-page search (150ms debounce), theme toggle, project search, card/list view
  • Zero-Config — sensible defaults; point at a JSONL file and go
  • 100% Local — pure offline tool; no AI used, no data sent anywhere, your transcripts stay on your machine

Installation

Via Homebrew (macOS)

brew install deepakkumardewani/weavr/weavr

Via cargo-binstall

cargo binstall weavr

Fetches the latest prebuilt binary from GitHub Releases. Falls back to building from source if no matching binary is found.

Via cargo install

cargo install weavr

Builds from source on crates.io. Requires Rust 1.80+.

Direct download

Prebuilt binaries are on the GitHub Releases page. Download the .tar.gz for your platform, extract, and place weavr on your PATH.

Platform Target triple
macOS (Apple Silicon) aarch64-apple-darwin
Linux (x86_64) x86_64-unknown-linux-gnu

Intel macOS has no prebuilt binary — install via cargo install weavr, or run the Apple Silicon build under Rosetta 2.

From source

git clone https://github.com/deepakkumardewani/weavr.git
cd weavr
cargo build --release
# binary is at target/release/weavr

Updating

weavr self-update       # update to the latest GitHub Release

A passive notice is printed when a newer version is available (throttled to once per 24 h). Set WEAVR_NO_UPDATE_CHECK=1 to disable.

Package-manager installs should use their native update commands:

brew upgrade weavr                  # Homebrew
cargo install --force weavr         # cargo install
cargo binstall --force weavr        # cargo-binstall

Requirements

  • Rust 1.80+ (source builds only; prebuilt binaries have no dependencies)

Usage

Basic export

weavr export session.jsonl
# writes session.html

weavr export session.jsonl -o /tmp/review.html --open-browser

weavr -i session.jsonl
# shorthand — equivalent to `export <INPUT>` (HTML only, full detail)

weavr -i session.jsonl --output-dir ~/desktop
# places session.html in ~/desktop/

--open-browser is a global flag — it works with all modes (export, -i, and --all-projects).

Example output:

Exported → ~/desktop/session.html

  Format:    HTML
  Messages:  106
  Tokens:    43,950 in / 110,380 out
  Took:      8ms

Token counts display with thousands separators and switch to M notation above one million (e.g. 1.23M, 12.5M).

Markdown export

weavr export session.jsonl --format md
# writes session.md

Detail levels:

weavr export session.jsonl --format md --detail full       # everything (default)
weavr export session.jsonl --format md --detail high       # messages + tool calls, no thinking
weavr export session.jsonl --format md --detail low        # messages only, no tool calls
weavr export session.jsonl --format md --detail minimal    # user + assistant text only
weavr export session.jsonl --format md --detail user-only  # user prompts only
Level Includes
user-only User prompts only — good input for an agent building a requirements doc
minimal User + assistant text
low + Key tool signals (WebSearch, WebFetch, Task delegations)
high + All tool calls; drops thinking blocks and system metadata
full Everything — thinking, system entries, all tool calls (default)

Feeding a past session to an LLM:

weavr export session.jsonl --format md --detail low --compact -o context.md

--compact strips timestamps, horizontal rules, and merges repeated same-type headings — significantly reducing token count.

All projects (batch)

weavr --all-projects
# walks ~/.claude/projects/ and generates weavr-out/
#   index.html
#   <project>/combined_transcripts.html
#   <project>/<session>.html

weavr --all-projects --projects-dir /path/to/projects --output-dir ./out

weavr --all-projects --no-individual-sessions
# skip per-session HTML; only index + combined pages

weavr --all-projects --open-browser
# opens index.html in the default browser when done

Example output:

Exporting 160 sessions across 10 projects → ./weavr-out

  ✓ my-project    49 sessions
  ✓ other-app     13 sessions
  ✓ index.html

Done: 162 sessions

  Format:    HTML (full)
  Messages:  26,289
  Tokens:    17.7M in / 11.1M out
  Took:      673ms

Filtering

By project:

weavr --all-projects --project cclog
# exports only the project whose name contains "cclog" (case-insensitive)

weavr --all-projects --project addy --output-dir ./out --open-browser
# partial match — "addy" matches "addyosmaniskills", "addy-osmani", etc.

--project does a case-insensitive substring match against both the internal directory name and the display name. Multiple projects can match. Exits with an error if nothing matches.

By date:

weavr export session.jsonl --from-date yesterday --to-date today
weavr --all-projects --from-date 2025-06-01 --to-date 2025-06-30

Accepts: today, yesterday, last week, last month, and ISO dates (YYYY-MM-DD). Sessions whose timestamp range doesn't overlap the filter window are skipped.

By session ID:

weavr --all-projects --session-id 6162c547
# exports only the matching session (prefix match)

Pagination

weavr export session.jsonl --page-size 50
# splits into session-page-1.html, session-page-2.html, ...

--page-size N sets messages per page. The first page includes the full chrome; subsequent pages are content-only.

Cache & output control

weavr --all-projects --no-cache      # skip cache entirely
weavr --all-projects --clear-cache   # drop and rebuild cache

weavr --all-projects --clear-output   # delete output dir before writing

weavr --debug export session.jsonl   # enable tracing output
weavr --debug --all-projects         # verbose logging for all-projects mode

Output Formats

HTML

  • Warm-neutral light/dark theme with embedded design tokens
  • Tool calls rendered as collapsible cards with syntax highlighting
  • Unified diff view for Edit / MultiEdit tool calls
  • Token usage displayed per message and as a session total
  • Thinking blocks in expandable sections
  • 100% self-contained — verified by a CI test that rejects any http(s):// URL in the output

Markdown

  • GitHub-Flavored Markdown compatible
  • Tool calls collapse to fenced code blocks with the tool name and key parameters
  • Diffs rendered as ```diff blocks with unified +/- hunks
  • --compact and --detail work orthogonally

Performance

Benchmarked with hyperfine (warmup + multiple runs, no cache) on Apple Silicon against claude-code-log over the same input. Re-run any time with just bench.

Mode weavr (Rust) claude-code-log (Python) Speedup
All projects (160 sessions, 97 MB) 1.32 s 24.10 s 18.2×
Single project (42 sessions) 466 ms 9.68 s 20.8×
Single session (19 MB, ~500 msgs) 27.5 ms 1.28 s 46.5×

Full methodology and optimization details in agent_docs/weavr-bench-results.md.

Roadmap

Done (v1.0)

  • Single-session HTML export with rich tool rendering
  • Markdown export with detail levels (full/high/low/minimal/user-only) + --compact
  • Project hierarchy + master index + per-project combined pages
  • SQLite cache for fast incremental rebuilds
  • Material 3 light/dark theme with flat dot-timeline
  • Client-side interactivity — filter chips, in-page search, theme toggle, sidebar nav
  • Date-range filtering with natural language (today, yesterday, last week)
  • Pagination for long sessions
  • Self-contained output (zero external URLs, CI-enforced)
  • Multi-channel install (brew, binstall, cargo) + self-update

Done (post-v1.0)

  • Per-project export — --project <name> filters batch export to matching projects (case-insensitive partial match)
  • --output-dir with -i — single-session shorthand now places output in the specified directory
  • Global --open-browser — works with export, -i, and --all-projects (opens index.html in batch mode)
  • Improved CLI output — colored progress (✓ per project), elapsed time (Took:), M-notation for large token counts, aligned stats block consistent across all modes

Planned

  • Interactive TUI — terminal-based transcript browser
  • Interactive timeline — zoomable, time-grouped view of message activity
  • Image rendering — inline display of images embedded in transcripts
  • JSON export — --format json for programmatic consumption
  • Windows support — currently macOS and Linux only
  • Cost estimation — per-session API cost breakdown
  • Live HTML sharing — deploy generated output to a hosted URL for one-click viewing and sharing in the browser

License

MIT