惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

WordPress大学
WordPress大学
A
About on SuperTechFans
量子位
B
Blog RSS Feed
让小产品的独立变现更简单 - ezindie.com
让小产品的独立变现更简单 - ezindie.com
博客园_首页
MongoDB | Blog
MongoDB | Blog
小众软件
小众软件
Blog — PlanetScale
Blog — PlanetScale
Microsoft Azure Blog
Microsoft Azure Blog
V
V2EX
Google DeepMind News
Google DeepMind News
Cyber Security Advisories - MS-ISAC
Cyber Security Advisories - MS-ISAC
H
Hackread – Cybersecurity News, Data Breaches, AI and More
G
Google Developers Blog
U
Unit 42
D
DataBreaches.Net
博客园 - Franky
D
Docker
宝玉的分享
宝玉的分享
Y
Y Combinator Blog
月光博客
月光博客
freeCodeCamp Programming Tutorials: Python, JavaScript, Git & More
Hugging Face - Blog
Hugging Face - Blog

Hacker News - Newest: "LLM"

GitHub - lechmazur/position_bias: A benchmark for testing whether LLM judges keep the same preference when two lightly edited versions of the same story are shown in opposite orders. Flex routing (EU and EFTA) Dark Factories: Retooling for LLM Velocity Ask HN: What would be the impact of a LLM output injection attack? GitHub - Oaklight/llm-rosetta: Production-ready LLM API translation layer for Python — bidirectional conversion between OpenAI, Anthropic & Google formats via hub-and-spoke IR. Optional API gateway. Streaming & non-streaming. Zero core deps. Contributions welcome! GitHub - browser-use/browser-harness: Self-healing browser harness that enables LLMs to complete any task. GitHub - moeen-mahmud/remen: Remen turns thoughts into something you can return to Analyzing 156 LLM Launch Posts on Hacker News ChatGPT vs Gemini vs Claude: The Best LLM Subscription You Should Buy GitHub - salaamalykum/quran-semantic-search: High-density RAG Semantic Search Engine & Quran Corpus (GEO/SEO Architecture) GitHub - NVIDIA/TensorRT-LLM: TensorRT LLM provides users with an easy-to-use Python API to define Large Language Models (LLMs) and supports state-of-the-art optimizations to perform inference efficiently on NVIDIA GPUs. TensorRT LLM also contains components to create Python and C++ runtimes that orchestrate the inference execution in a performant way. The State of LLM Bug Bounties in 2026 Operational Readiness Criteria for Tool-Using LLM Agents Meshcore: Architecture for a Decentralized P2P LLM Inference Network How an LLM becomes more coherent as we train it GitHub - seetrex-ai/laimark GitHub - Jossifresben/BibCrit: AI-assited biblical textual criticism GitHub - wastedcode/memex: File system based wiki, maintained by Claude 99helpers.com GitHub - cliver-project/AITrigram GitHub - unbody-io/adapt: A self-evolving memory layer for AI agents. GitHub - hb20007/awesome-gen-ai-fails: A list of incidents where reliance on generative AI and LLMs resulted in harm to companies, individuals, or society GitHub - nevenkordic/localmind: Run any local LLM with persistent memory and context. CLI agent over Ollama with SQLite-backed hybrid recall. No cloud. Ask HN: What are the machine requirements for a LLM like Llama-3.1-8B? Faster LLM Inference via Sequential Monte Carlo grpo explained: group relative policy optimization for llm finetuning - cgft Stop comparing price per million tokens: the hidden LLM API costs · TensorZero Andrej Karpathy's LLM Wiki Is a Bad Idea GitHub - GG-QandV/mnemostroma: Offline RAM-first cognitive leer/coprocessor for AI agents and robotics. Solves "Context Abandonment" with 20-80ms latency using a dual-thread biomimetic memory architecture (ONNX + SQLite WAL). mempalace/agent at agent · skorotkiewicz/mempalace
GitHub - MTimma/knowerage: Local MCP server that tracks A...
mtimma · 2026-04-28 · via Hacker News - Newest: "LLM"

Knowerage project icon

Links: GitHub · Glama MCP listing · npm @mtimma/knowerage

Quick Start

Requirements: Node.js 18 or newernpx must be on your PATH (it comes with npm, which is included with Node).

MCP server configuration

Register Knowerage wherever your MCP host expects server definitions (for example some clients use .cursor/mcp.json or .vscode/mcp.json; others use environment variables or a UI—follow your host’s documentation). Use the same server entry shape:

{
  "mcpServers": {
    "knowerage": {
      "command": "npx",
      "args": ["@mtimma/knowerage"],
      "env": {
        "KNOWERAGE_WORKSPACE_ROOT": "${workspaceFolder}",
        "KNOWERAGE_AUTO_FULL_RECONCILE": "true"
      }
    }
  }
}

Replace ${workspaceFolder} with your project root if your host does not expand that variable.

KNOWERAGE_AUTO_FULL_RECONCILE is optional: when unset, empty, or not a truthy value, the file watcher defaults to off. Set to 1, true, yes, or on (trimmed, case-insensitive) to enable. When on, the server watches knowerage/ and, after a short debounce, runs knowerage_reconcile_all on filesystem changes. That is not the same as running a full reconcile after every MCP tool call—it only reacts to file changes under knowerage/. Registry writes to registry.json are ignored by the watcher so saves do not loop.

How to use Knowerage

After the MCP server is configured, you talk to your assistant in normal sentences. You do not need to memorize tool names.

Analyse or document code

Point at files, classes, or behaviour you care about. For example:

  • Using Knowerage, analyse the logical algorithm workflow in main.java.
  • Analyze the data entity reconciliation and versioning logic in the ETL service.

The assistant creates or updates markdown under knowerage/analysis/ and records coverage in knowerage/registry.json (see How It Works below).

Coverage and gaps (same project, later chat or another agent)

When you already have analyses in the tree, you can ask:

  • In percentage, how much of the code has our analysis covered?
  • What part of this codebase is not yet analysed?

Knowerage answers these from the registry and coverage helpers (for example overview, per-file status, and stale lists)—not from hand-waving over the repo.

Alternative approaches

Install via npm

npx @mtimma/knowerage

Or build from source

cargo build --release
./target/release/knowerage-mcp

How It Works

  1. AI agent creates analysis .md files with YAML frontmatter declaring source file and covered line ranges
  2. Registry (knowerage/registry.json) tracks analysis records with SHA-256 hashes for freshness
  3. MCP tools expose create, reconcile, query, and export operations
  4. Agent says "analyze X" → full workflow runs automatically (create → reconcile → record)

Registry file shape (knowerage/registry.json)

The on-disk format is a JSON object whose keys are analysis paths (strings). Each value is one record (see contracts/contracts.md). A full sample with two records lives at examples/registry.sample.json.

flowchart TB
  subgraph file["knowerage/registry.json"]
    O["Top-level JSON object"]
    O --> K["Each key: analysis markdown path, e.g. knowerage/analysis/.../topic.md"]
    K --> V["Value: one RegistryRecord"]
  end

  subgraph rec["RegistryRecord fields"]
    ap["analysis_path · source_path"]
    cr["covered_ranges: [[start,end], ...]"]
    h["analysis_hash · source_hash (sha256:… )"]
    t["record_created_at · record_updated_at (ISO 8601)"]
    st["status: fresh | stale_doc | stale_src | missing_src | dangling_doc"]
  end

  V --> rec
Loading

Frontmatter for analysis .md files is specified separately in the contracts doc (metadata schema), not inside registry.json.

MCP Tools

Tool Purpose
knowerage_create_or_update_doc Create/update analysis document
knowerage_parse_doc_metadata Parse and validate frontmatter
knowerage_reconcile_record Reconcile one analysis record
knowerage_reconcile_all Full rescan/rebuild
knowerage_get_file_status Analyzed vs missing ranges
knowerage_list_stale List stale/problematic records
knowerage_list_registry Full registry snapshot (same shape as registry.json, sorted keys)
knowerage_get_tree Tree/grouped coverage
registry_export_report Export snapshot (JSON/YAML/TXT/HTML)
knowerage_generate_bundle Chunked export of selected analyses (toc*.md, combined*.md, manifest.json)

Project Structure

knowerage/                  # Created per-project
├── analysis/              # Analysis markdown files
│   └── **/*.md
└── registry.json          # Coverage registry

src/                       # Rust MCP server
├── main.rs
├── lib.rs
├── types.rs
├── parser.rs
├── registry.rs
├── mcp.rs
├── security.rs
└── export.rs

Documentation

Security

  • All paths validated against workspace root
  • Path traversal (..) rejected
  • Atomic writes for registry (crash-safe)
  • No secrets in analysis files or reports
  • SHA-256 hash-based freshness (survives git pull)

License

MIT — copyright Martins Timma.

Parts of this project were written or refined with generative AI coding assistants. Human review applies to design, security-sensitive behavior, and releases.