Skip to content

CHANGELOG.md

Latest commit

 

History

History
92 lines (66 loc) · 3.12 KB

CHANGELOG.md

File metadata and controls

92 lines (66 loc) · 3.12 KB

Changelog

All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.

[Unreleased]

Added

Changed

Fixed

[0.5.0] - 2025-10-24

Added

  • Sentry LLM Monitoring Support: First-class integration with Sentry's AI Performance monitoring
    • sentry_adapter.py following Logfire pattern (native SDK config, OTEL API execution)
    • Automatic configuration via SENTRY_DSN environment variable
    • gen_ai.* attributes for Sentry's AI Monitoring dashboard
    • Full integration with Sentry's error tracking + LLM monitoring
    • Token usage tracking, cost analysis, and tool execution visibility
    • Comprehensive test suite with 91% coverage
    • Complete documentation in README with setup instructions

[0.2.1] - 2025-10-15

Fixed

  • Critical: Fixed CLI entry point to call app() instead of main()
    • v0.2.0 had a bug where all commands entered interactive mode
    • --version, --print, and other flags now work correctly
    • Issue only affected installed package (not python -m usage)

[0.2.0] - 2025-10-15

Changed

  • Major CLI Refactor: Simplified argument parsing using Typer's variadic arguments
    • Reduced parsing logic from ~200+ lines to ~50 lines
    • Removed complex custom prompt extraction that was prone to bugs
    • Simple heuristic: last non-option argument is the prompt
    • Proper handling of --flag=value, --flag value, and --flag (boolean) formats
  • Improved test suite with proper mocking following pytest standards
    • Added 20 new CLI parsing tests
    • Mock external dependencies instead of calling Claude CLI during tests
    • Faster test execution (0.30s for full suite)

Fixed

  • Fixed argument parsing bugs where flags were incorrectly treated as prompts
  • Fixed test for KeyboardInterrupt handling to properly expect typer.Exit

[0.1.2] - 2025-01-XX

Added

  • Initial public release! 🎉
  • OpenTelemetry instrumentation for Claude agents
  • Comprehensive telemetry capture (prompts, tool calls, tokens, costs)
  • Logfire integration with LLM-specific UI features
  • Support for any OTEL-compatible backend
  • CLI tool (claudia) for quick agent execution
  • Interactive and non-interactive agent modes
  • MCP server support (HTTP and stdio)
  • Comprehensive test suite
  • Pre-commit hooks for code quality
  • CI/CD pipeline with GitHub Actions

Features

  • 🤖 Captures every prompt, tool call, token count, and cost as structured OTEL spans
  • 📊 Works with Logfire, Datadog, Honeycomb, Grafana, and any OTEL collector
  • 🔧 Hook-based telemetry (no monkey-patching required)
  • 💡 Enhanced Logfire features with LLM-specific UI tagging
  • 🎨 Beautiful console output with emoji indicators
  • ⚡ Async-first API with sync convenience wrappers
  • 🔌 Extensible and maintainable architecture

Release History

Note: This project uses setuptools-scm for automatic version management from git tags.