feat: update README and feature comparison chart for clarity and accuracry #201
Conversation
|
👋 Hey @bashandbone, Thanks for your contribution to codeweaver! 🧵You need to agree to the CLA first... 🖊️Before we can accept your contribution, you need to agree to our Contributor License Agreement (CLA). To agree to the CLA, please comment:
Those exact words are important1, so please don't change them. 😉 You can read the full CLA here: Contributor License Agreement ✅ @bashandbone has signed the CLA. You can retrigger this bot by commenting recheck in this Pull Request. Posted by the CLA Assistant Lite bot. Footnotes
|
Reviewer's GuideThis PR overhauls the README and competitive comparison docs to position CodeWeaver as "semantic code search for Claude," updates competitive metrics and terminology (e.g., language-aware chunking, provider counts), clarifies installation/MCP configuration and offline operation, and removes outdated Claude-specific docs/configs while keeping generator scripts and metadata in sync with the new language and capabilities. File-Level Changes
Tips and commandsInteracting with Sourcery
Customizing Your ExperienceAccess your dashboard to:
Getting Help
|
There was a problem hiding this comment.
Pull request overview
This pull request updates CodeWeaver's README and feature comparison documentation to improve clarity and competitive positioning. The changes focus on repositioning CodeWeaver as a semantic code search tool for Claude, adding comprehensive competitive analysis including a new comparison with Serena, and streamlining the documentation structure.
Key Changes:
- Rewrote the README introduction to emphasize semantic code search capabilities and concrete use cases
- Added detailed CodeWeaver vs. Serena competitive analysis with practical tradeoffs
- Updated feature comparison matrix to include Serena and revised metrics across all competitors
- Streamlined README by removing extensive sections on architecture, roadmap, and project details in favor of focused value proposition
Reviewed changes
Copilot reviewed 2 out of 2 changed files in this pull request and generated 9 comments.
| File | Description |
|---|---|
| README.md | Complete overhaul of introduction and positioning; new tagline emphasizing "semantic code search for Claude"; simplified feature grid; removed extensive documentation sections; updated configuration examples |
| claudedocs/codeweaver_feature_comparison_chart.md | Added Serena to comparison matrix; new detailed CodeWeaver vs. Serena breakdown section; updated metrics and feature counts; revised industry context; streamlined performance section |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
README.md
Outdated
| # CodeWeaver | ||
|
|
||
| ### The missing abstraction layer between AI and your code | ||
| ### Semantic code search for Claude — across 170+ languages |
There was a problem hiding this comment.
The tagline mentions "170+ languages" but the feature comparison table below shows "166+" languages. This inconsistency should be resolved - use the same number in both places.
| ### Semantic code search for Claude — across 170+ languages | |
| ### Semantic code search for Claude — across 166+ languages |
README.md
Outdated
| - **Hierarchical merging** | ||
| - **Environment overrides** | ||
| ### 🔌 Provider Flexibility | ||
| - **17+ embedding providers** |
There was a problem hiding this comment.
The claim "17+ embedding providers" in this section contradicts other parts of the document which state "16+ embedding providers" (line 72, 178, and line 327 in the feature comparison chart). Ensure consistency across all mentions.
| - **17+ embedding providers** | |
| - **16+ embedding providers** |
| 1. Embedding provider diversity (20+ vs 1-5) | ||
| 1. Embedding provider diversity (16+ vs 1-5) | ||
| 2. Hybrid search (dense + sparse by default) | ||
| 3. Language support breadth (170+ with semantic-like chunking) |
There was a problem hiding this comment.
The feature comparison chart states "Language support breadth (170+ with semantic-like chunking)" but earlier in the document the numbers vary between 166+ and 170+. This should be consistent throughout the document.
| | **Tool Count** | **1** | **17** | N/A | N/A | N/A | N/A | N/A | N/A | | ||
| | **Prompt Overhead** | **~500 tokens** | **~17,000 tokens** | N/A | N/A | N/A | N/A | N/A | N/A | | ||
| | **Search Speed** | Moderate (embeddings) | **Very fast (LSP)** | Moderate | Server-side | Fast | Moderate | Fast | On-demand | | ||
| | **Embedding Providers** | **16+** | 0 (no embeddings) | 1-2 | 1 | 0 (deprecated) | 4-5 | 1 | 0 | |
There was a problem hiding this comment.
The embedding provider count in this table shows "16+" but in the Detailed Feature Breakdown section (line 318) it says "16+ providers" while the conclusion (line 327) mentions provider diversity as "(16+ vs 1-5)". However, the previous version mentioned "20+ providers" which was changed to "16+" in some places but not all. Ensure the count is accurate and consistent - if it's 16+, update all references; if it's actually more, use the correct number.
README.md
Outdated
| ### 🔌 Provider Flexibility | ||
| - **17+ embedding providers** | ||
| - **50+ embedding models** | ||
| - **Sparse & dense**embedding model support |
There was a problem hiding this comment.
Typo: "Sparse & dense**embedding model support" is missing a space before "embedding". Should be "Sparse & dense embedding model support".
| - **Sparse & dense**embedding model support | |
| - **Sparse & dense** embedding model support |
README.md
Outdated
| [arch_find_code]: <src/codeweaver/agent_api/find_code/ARCHITECTURE.md> "find_code Architecture" | ||
| [architecture]: <ARCHITECTURE.md> "Overall Architecture" | ||
| [bashandbone]: <https://github.com/bashandbone> "Adam Poulemanos' GitHub Profile" | ||
| [competitive_analysis]: <src/codeweaver/claudedocs/competitive_feature_comparison_chart.md> "See how CodeWeaver stacks up" |
There was a problem hiding this comment.
The URL for the competitive analysis link appears to be incorrect. It references "src/codeweaver/claudedocs/competitive_feature_comparison_chart.md" but the actual file in the PR is "claudedocs/codeweaver_feature_comparison_chart.md" (not "competitive_feature_comparison_chart.md"). This link will be broken.
| [competitive_analysis]: <src/codeweaver/claudedocs/competitive_feature_comparison_chart.md> "See how CodeWeaver stacks up" | |
| [competitive_analysis]: <claudedocs/codeweaver_feature_comparison_chart.md> "See how CodeWeaver stacks up" |
…sistency and fairness across docs and comparisons. Updated all docs accordingly. Cleaned up older planning docs
|
Review the following changes in direct dependencies. Learn more about Socket for GitHub.
|
![sourcery-ai[bot]](https://avatars.githubusercontent.com/in/48477?s=60&v=4){kind=small}
There was a problem hiding this comment.
Hey there - I've reviewed your changes - here's some feedback:
- The Quick Reference Matrix and several competitive stats (e.g., provider counts, language counts, prompt overhead) are now duplicated between the README and docs/comparison.md; consider centralizing these in a single source (or generating them from a script/constant) to avoid future drift.
- Numbers like
17 embedding providers,166+ languages, and token counts are hard-coded throughout the comparison doc and README while some of these values are already available programmatically (e.g., len(_languages()), embedding_providers()); wiring the docs to those programmatic sources would make them more robust as the product evolves. - The new Serena comparison introduces specific measurements (tool counts, token overhead, language counts) that may change quickly as clients update; consider either linking to a methodology/measurement section or parameterizing these values to make it easier to keep them current.
Prompt for AI Agents
Please address the comments from this code review:
## Overall Comments
- The Quick Reference Matrix and several competitive stats (e.g., provider counts, language counts, prompt overhead) are now duplicated between the README and docs/comparison.md; consider centralizing these in a single source (or generating them from a script/constant) to avoid future drift.
- Numbers like `17 embedding providers`, `166+ languages`, and token counts are hard-coded throughout the comparison doc and README while some of these values are already available programmatically (e.g., len(_languages()), embedding_providers()); wiring the docs to those programmatic sources would make them more robust as the product evolves.
- The new Serena comparison introduces specific measurements (tool counts, token overhead, language counts) that may change quickly as clients update; consider either linking to a methodology/measurement section or parameterizing these values to make it easier to keep them current.Help me be more useful! Please click 👍 or 👎 on each comment and I'll use the feedback to improve your reviews.
There was a problem hiding this comment.
Pull request overview
Copilot reviewed 24 out of 25 changed files in this pull request and generated no new comments.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
Co-authored-by: Copilot <[email protected]> Signed-off-by: Adam Poulemanos <[email protected]>
There was a problem hiding this comment.
Pull request overview
Copilot reviewed 24 out of 25 changed files in this pull request and generated no new comments.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
…, WHY.md, and comparison chart
…ng experimental note and fixing formatting
There was a problem hiding this comment.
Pull request overview
Copilot reviewed 25 out of 26 changed files in this pull request and generated 1 comment.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
docs/WHY.md
Outdated
| - Even well-meaning MCP tools contribute: several popular MCP servers have **16,000+ tokens in prompt overhead** -- all the prompts they supply **every single message** to tell an agent about their available tools and how to use them [^1] | ||
|
|
||
| CodeWeaver takes a different approach: | ||
| CodeWeaver strives to approac: |
There was a problem hiding this comment.
Typo: "approac" should be "approach"
| CodeWeaver strives to approac: | |
| CodeWeaver strives to approach: |
…and enhancing readability
There was a problem hiding this comment.
Pull request overview
Copilot reviewed 25 out of 26 changed files in this pull request and generated no new comments.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
This pull request significantly revises the
README.mdand updates the feature comparison chart for CodeWeaver, focusing on clearer messaging, improved competitive positioning, and more actionable guidance for users. The most important changes include a complete rewrite of the introduction and feature sections to emphasize semantic code search for Claude, a streamlined competitive analysis, and clearer setup instructions. Additionally, the feature comparison chart now includes Serena and provides a more nuanced breakdown of strengths and tradeoffs.Major content and documentation updates:
README overhaul and positioning
.mcp.jsonexamples for both stdio and HTTP setups.Feature and capability clarification
Competitive analysis improvements:
These changes make the documentation more user-focused, clarify CodeWeaver’s unique value, and provide actionable guidance for both setup and tool selection.
Summary by Sourcery
Revise positioning and documentation to emphasize CodeWeaver as semantic code search for Claude, and update competitive materials and terminology for more accurate feature and language-support descriptions.
Enhancements:
Documentation:
Chores: