Skip to content

CHANGELOG.md

Latest commit

 

History

History
281 lines (217 loc) · 12.9 KB

CHANGELOG.md

File metadata and controls

281 lines (217 loc) · 12.9 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]

[0.4.1] - 2026-01-20

Fixed

  • Colocation in monorepos now resolves paths relative to the target package directory, so .figma.ts(x) files land next to the actual component files (affects Chakra, ZapUI, etc.)

[0.3.2] - 2026-01-14

Changed

  • Default concurrency reduced from 8 to 5 for better compatibility with LLM proxies
  • Improved TOML comments for concurrency setting to explain its purpose and troubleshooting

Fixed

  • Added comprehensive error handling for common API errors:
    • 402 billing/quota errors with links to billing dashboards
    • 408 timeout errors with retry guidance
    • 413/context length errors with model suggestions
    • 502/503 service unavailable with concurrency advice (helps LiteLLM/Bedrock users)
    • 529 overloaded (Anthropic-specific) with wait guidance
    • Content policy violations with troubleshooting tips
  • Rate limit (429) errors now suggest lowering concurrency as first troubleshooting step

[0.3.1] - 2026-01-14

Changed

  • Renamed intermediate results directory from superconnect/ to superconnect-logs/ for clarity
    • All pipeline artifacts (repo-summary.json, figma-components-index.json, orientation.jsonl, codegen-summaries/, etc.) now written to superconnect-logs/
    • Updated documentation, scripts, tests, and .gitignore to reflect new naming
    • No changes to config file name (superconnect.toml) or package/binary names

[0.3.0] - 2026-01-13

Changed

  • Agent SDK is now the default backend — 33% faster, 6x fewer tokens vs Messages API
  • Default model changed to claude-sonnet-4-5
  • Default max tokens increased from 2048 to 16384

Added

  • Agent SDK mode with autonomous codebase exploration (Read, Glob, Grep tools)
  • --no-agent-sdk flag to use Messages API instead
  • Expanded E2E test coverage (7 Chakra + 6 ZapUI components with semantic assertions)

Fixed

  • Validation false positives on valid Code Connect files
  • Angular codegen confusion when React quickstart docs present
  • Agent SDK thinking text leaking into generated code

Internal

  • Migrated from Jest to node:test (faster, zero deps)
  • Consolidated utilities into src/util/naming.js and src/util/fs-helpers.js
  • Streamlined documentation

[0.2.10] - 2025-12-18

Fixed

  • Windows PowerShell output visibility: Child process output (Figma scan details, code generation progress) now displays correctly on Windows
    • Root cause: spawnSync with shell: true spawned cmd.exe as intermediary, which interfered with console output and ANSI color codes
    • Solution: Call Node.js scripts directly via process.execPath with argument arrays instead of shell command strings
    • Eliminates shell quoting/escaping issues and works identically across all platforms

[0.2.9] - 2025-12-18

Changed

  • Refactor validation layer to use hybrid AST/regex approach
    • Replace regex-based figma.*() extraction with ts-morph AST traversal for better edge case handling
    • Keep proven regex patterns for template/JSX validation
    • Improves correctness (+43%) and maintainability (+33%) with minimal performance impact

[0.2.8] - 2025-12-17

Performance

  • Replace npx with direct CLI invocation for Figma validation (4x faster on Windows)
    • Resolve @figma/code-connect binary path directly from node_modules
    • Use process.execPath (node) to invoke CLI instead of npx
    • Eliminates ~30s npx overhead per CLI call on Windows
    • Windows CI improved from 64s to 16s for unit tests
    • Reduce timeout from 120s to 30s (no package download overhead)

Changed

  • Trigger CI unit tests on version tags in addition to branch pushes
  • Improve release process documentation with pre-flight checks and error summaries
  • Remove global Figma CLI pre-install from CI workflow (no longer needed)

[0.2.7] - 2025-12-17

Fixed

  • CRITICAL: Windows compatibility for validation - add shell: true to all npx spawn calls
    • v0.2.6 used npx.cmd but still failed with EINVAL on Windows
    • Now uses shell: true which is the Node.js recommended approach for cross-platform compatibility
    • Removes platform-specific npx.cmd detection in favor of simpler, more robust solution

[0.2.6] - 2025-12-17

Fixed

  • CRITICAL: Windows npx compatibility - use npx.cmd on Windows to prevent ENOENT errors
    • Previously caused 100% validation failure on Windows with "spawnSync npx ENOENT"
    • Affects all Code Connect file validation attempts
    • Now correctly detects Windows platform and uses npx.cmd

Changed

  • Streamline agent documentation for clarity and reduced token usage

[0.2.5] - 2025-12-17

Fixed

  • CRITICAL: Move @figma/code-connect from devDependencies to dependencies
    • Previously, global npm installs (npm install -g @bitovi/superconnect) did not include the Figma CLI
    • This caused 100% validation failure on all Code Connect files with "unknown error"
    • Now the CLI is always installed with superconnect
  • Add upfront check for Figma CLI availability with clear troubleshooting steps
  • Improve validation error messages to show exit code, stdout, and stderr for debugging
  • Fix Anthropic SDK 0.71.2 compatibility:
    • Add explicit stream: false for non-streaming requests
    • Increase timeout to 20 minutes for long-running orientation tasks (SDK default was 10 minutes)

Changed

  • Update dependencies to latest compatible versions:
    • @anthropic-ai/sdk: 0.18.0 → 0.71.2
    • commander: 12.1.0 → 14.0.2
    • openai: 4.104.0 → 6.14.0
    • undici: 6.18.1 → 7.16.0
    • @figma/code-connect: 1.3.12 (already current)
  • Keep chalk at 4.1.2 and p-limit at 3.1.0 (newer versions are ESM-only, incompatible with CommonJS)

[0.2.4] - 2025-12-17

Fixed

  • Add helpful error messages when API returns "Invalid model name" errors (400 status)
    • Shows current model being used
    • Suggests common alternatives (gpt-4o, claude-sonnet-4-5, etc.)
    • Explains how to set model via superconnect.toml or CLI flag
    • Links to model documentation
  • Warn users when using custom base_url without explicitly setting a model
    • Prevents confusion when default model doesn't exist on custom endpoints
    • Applies to LiteLLM, Azure OpenAI, vLLM, and other OpenAI-compatible proxies

[0.2.3] - 2025-12-17

Fixed

  • Windows compatibility: Use Node's native fetch instead of undici to fix "fetch failed" errors on Windows PowerShell with corporate networks
  • Include CHANGELOG.md in npm package files array so users can see release notes
  • Change CI workflow to use npm ci instead of npm install for consistency with publish workflow
  • Add .npmrc with engine-strict=true to enforce Node >=22 requirement (was only advisory before)
  • Remove redundant .npmignore file (files[] array in package.json already controls what's published)

[0.2.2] - 2025-12-17

Fixed

  • Include .version-sha in npm package files array (was being generated but not included in published package)

[0.2.1] - 2025-12-17

Added

  • Include git SHA in npm package version output via prepublishOnly script
  • E2E tests now run in parallel using GitHub Actions matrix strategy

Fixed

  • Read api_key from superconnect.toml when validating agent token (interactive setup bug where custom API keys were written to TOML but not read during validation)

[0.2.0] - 2025-12-17

Changed

  • BREAKING: Rename [agent] config keys for semantic clarity:
    • backend → api (values: "anthropic" or "openai")
    • sdk_model → model
    • Value "claude" → "anthropic" (reflects API format, not model)
  • BREAKING: Rename CLI flags: --agent-backend → --agent-api
  • Old config keys still work with deprecation warnings; update recommended
  • Make superconnect.toml self-documenting with clearer comments and actionable guidance
  • Replace all gpt-4 references with gpt-5.1-codex-mini (only gpt-5 class models in examples)

Added

  • LiteLLM and OpenAI-compatible endpoint support for agent APIs
  • New base_url and api_key fields in [agent] section of superconnect.toml for custom endpoints
  • Support for OPENAI_BASE_URL environment variable
  • CLI flags --agent-base-url and --agent-api-key for run-orienter.js and run-codegen.js
  • --version flag to display version number with git SHA (e.g., "0.2.0 (abc1234)")
  • Interactive prompting for custom endpoint configuration (base_url/api_key) when choosing OpenAI API on first run
  • All configuration options now visible in generated superconnect.toml with helpful comments
  • Deprecation warnings when old config keys (backend, sdk_model) are detected
  • Comprehensive unit tests for custom endpoint support
  • Documentation for using LiteLLM, Azure OpenAI, vLLM, and other OpenAI-compatible servers

Fixed

  • Enhance error diagnostics for network and certificate issues in corporate environments across all pipeline stages
  • Add detailed error logging to agent adapters (error type, code, status, cause) to help diagnose connection failures
  • Improve error messages for TLS/SSL certificate errors with specific troubleshooting steps for locked-down networks
  • Add network error detection and corporate environment guidance to Figma scan stage
  • Add enhanced authentication and file-not-found error messages to Figma API requests
  • Preserve network error details in React/Angular codegen (no longer wrapped as generic "Agent error")
  • Flag network vs agent errors in codegen attempt logs for better diagnostics
  • Always display log file locations when stages fail so users can access complete diagnostic information
  • Point users to docs/NETWORK-TROUBLESHOOTING.md for detailed corporate network help

[0.1.10] - 2025-12-16

Changed

  • Improve README comparison with Figma Code Connect, clarifying that Superconnect works fully automatically and supports Angular/HTML (which Figma's interactive setup does not)

[0.1.9] - 2025-12-16

Fixed

  • Add scope: '@bitovi' parameter to setup-node in publish workflow for proper scoped package authentication

[0.1.7] - 2025-12-15

Fixed

  • Remove prepublishOnly hook that caused npm publish to run tests twice, leading to intermittent publish failures

[0.1.6] - 2025-12-15

Fixed

  • Fix duplicate logDir property in test file causing intermittent test failures

[0.1.5] - 2025-12-15

Changed

  • Upgrade project to require Node.js ≥ 22.0.0 (previously ≥ 20.0.0)
  • Update CI/CD workflows to use Node 22 consistently
  • Update documentation (README.md, TESTING.md) to reflect Node 22 requirement

0.1.4 - 2025-12-15

Fixed

  • Fix test failures in CI by adding missing logDir parameter to processComponent calls
  • Suppress Anthropic SDK --localstorage-file warnings in tests with NODE_NO_WARNINGS=1

0.1.3 - 2025-12-15

Fixed

  • Fix DEP0190 Node.js deprecation warning by removing shell:true from spawnSync calls with args arrays
  • Fix boolean variant validation to recognize True/False, Yes/No, On/Off patterns as valid for figma.boolean()
  • Strengthen prompt guidance prohibiting &&, ||, and ternary operators in example JSX/templates to prevent Code Connect validation errors

Changed

  • Enhance retry prompts with explicit CRITICAL section directing agent to use only properties from Figma Component Data
  • Add comprehensive agent I/O transcript logging for debugging (system/user prompts, outputs, token usage)
  • Rename output directories for clarity: codegen-logs → codegen-summaries, mapping-agent-logs → codegen-agent-transcripts

0.1.1 - 2025-12-15

Changed

  • Update repository URLs to github.com/bitovi/superconnect
  • Set up automated release workflow with GitHub Actions
  • Add RELEASE.md documenting the release process
  • Update documentation for npm granular access tokens

0.1.0 - 2025-12-15

Added

  • Initial release of @bitovi/superconnect
  • AI-powered code generation for Figma Code Connect mappings
  • Support for React and Angular frameworks
  • Five-stage pipeline: repo summarizer, Figma scan, orienter, codegen, finalizer
  • Claude (Anthropic) and OpenAI backend support
  • Direct codegen with built-in validation and retry logic
  • Graceful interrupt handling and partial runs
  • Comprehensive test suite with E2E validation against Chakra UI and ZapUI

Features

  • Automatic framework detection (React/Angular)
  • Component-level filtering with --only and --exclude flags
  • Concurrent codegen with configurable parallelism
  • Cached system prompts for efficient LLM usage
  • Detailed logging and colorized console output
  • figma.config.json generation for Code Connect CLI
  • Support for monorepo path aliases
  • Document URL substitutions for Figma node links