Your Codebase, Understood.

Context infrastructure for AI.

We solve the hardest problem in AI development: making sure your AI actually understands your work. Semantic search, automatic context routing, and intelligent memory—across any LLM.

100% Local Free Tier | Claude Code | Cursor | Windsurf | Adaptive Context
terminal — ~/my-project
> |
[pyckle] route: calibration_flow → semantic [0.91]
[pyckle] graph: +2 neighbors via calibration.py
[pyckle] 5 files · 612 tokens injected · 8ms
[pyckle] calibration.py:187 · cloud_router.py:44 · config.py:23 · adaptive_threshold.py:61 · search.py:112

The calibration threshold is updated in three places:

1. update_threshold() in calibration.py:187 is called after every miss — it lowers the similarity floor by the configured step size.

2. AdaptiveRouter.recalibrate() in cloud_router.py:44 reads the new threshold and re-weights candidate scoring on the next query.

3. The floor and ceiling bounds live in config.py:23 as CALIBRATION_MIN and CALIBRATION_MAX — they prevent runaway drift.

With Pyckle: 0 tokens Without: 0 tokens
90% saved

You ask a question. Pyckle handles the rest.

Simulated output. Token savings vary by codebase and query.

The Problem

AI forgets. Every time.

Up to 87%

of tokens can be wasted on irrelevant context

Up to 10x

more time spent hunting for files without context

0%

memory between sessions

Every AI coding tool uses the same models. The difference is context.
We make context automatic.

Our Approach

Context that works everywhere

Semantic Search

Find code by meaning, not keywords. Ask "where do we handle auth?" and get the right files—not string matches.

Auto-Context Routing

Every prompt gets the right code snippets automatically. No manual @mentions. No wasted tokens. ~50ms typical latency.

Local-First

Your code stays on your machine. Free tier runs entirely local—no API keys, no cloud dependency, no trust required.

Architecture

How it works

Data Flow Pipeline

User Query
Pyckle MCP
Semantic Search
Ranked Context
AI

Session Warming States

Cold Start Warm Fully Warm
First query Context loaded Full memory

MCP Tools

22 tools, one protocol

Everything your AI needs to understand your codebase.

search_code

Semantic code search by meaning, not keywords.

Learn more →

index_codebase

Index your codebase for instant semantic queries.

Learn more →

index_stats

View indexing statistics and health metrics.

Learn more →

token_stats

Track token usage and optimize context budget.

Learn more →

session_continue

Resume sessions with full context memory.

Learn more →

session_summary

Generate summaries of session activity.

Learn more →

register_edit

Track file edits for context coherence.

Learn more →

graph_neighbors

Explore code dependencies and connections.

Learn more →

graph_impact

Analyze blast radius of code changes.

Learn more →

index_obsidian

Index an Obsidian vault for semantic retrieval.

Learn more →

index_notion

Index a Notion database or page for semantic queries.

Learn more →

index_git_history

Query commit history with natural language.

Learn more →

auto_context

Model-agnostic prompt routing for any MCP client.

Learn more →

autoloop_init

Start an autonomous goal-directed iteration session.

Learn more →

autoloop_log

Record an iteration result — keep, discard, or crash.

Learn more →

autoloop_status

Get live progress and metrics for an active loop.

Learn more →

autoloop_history

List all autoloop sessions for a codebase.

Learn more →

autoloop_complete

Mark an autoloop as complete and archive results.

Learn more →

index_git_issues

Index GitHub and GitLab issues for semantic search.

Learn more →

add_memory

Store a decision or insight in persistent memory.

Learn more →

search_memory

Retrieve past decisions and context from memory.

Learn more →

get_coverage

Map source files to their test coverage.

Learn more →

Products

Built for how you work

Context infrastructure for every AI workflow.

Pyckle Pro

Everything included. One price.

Semantic context for every AI prompt — invisible, ~50ms, no workflow changes. Works natively with Claude Code and any MCP-compatible editor.

  • Semantic code search & auto-context injection
  • MCP support — Cursor, Windsurf, Continue & more
  • Session memory & context routing
  • All features included — no add-ons
$10 /month
Learn More

Pyckle Embeddings

PyckLM-powered code vectors

Production-grade code embeddings via REST API. PyckLM understands code structure — not just text. Drop-in for any app that needs semantic code search.

  • Code-optimized embeddings, not generic text
  • REST API — drop into any stack
  • Hybrid semantic + keyword retrieval
  • Free tier — no credit card required
$10 /month
Learn More

Pyckle Router

The AI router that knows your code

Multi-provider AI routing with automatic codebase search. Route queries to the best model — Anthropic, OpenAI, Groq, Mistral, Ollama — with your codebase context pre-loaded automatically.

  • 7-provider routing with automatic failover
  • MCP tool loop — searches your codebase automatically
  • Cost-aware routing with configurable session budget
  • Local-first option — zero data egress via Ollama
$10 /month
Learn More

Clip

Your reading, made intelligent.

Save any web page in one click. The more you save, the smarter it gets — Clip builds a personal knowledge base from your reading habit, automatically.

  • Save anything, search everything
  • AI chat across your entire library
  • Knowledge graph — see how ideas connect
  • Chrome extension + Web app
Free to start
Learn More

Why Pyckle

The name says it all

Pyckle = preservation. In a world where AI systems constantly reset and forget, we're the constant.

We ensure your most valuable asset—your context—remains accessible, optimized, and alive. Whether you're using Claude, GPT, Gemini, or whatever comes next.

Ready to preserve your intelligence?

Works with Claude Code, Cursor, Windsurf, and any MCP-compatible client.