code-context-engine 0.4.3

Index your codebase. AI searches instead of re-reading files. Save 70%+ on tokens. Works with Claude Code, Cursor, VS Code, Gemini CLI, and Codex.

Code Context Engine

Index your codebase. AI searches instead of re-reading files. 93% token savings, benchmarked.

Works with your editor

           

One command. Index your codebase. Your AI coding agent searches instead of reading entire files.
Zero-cloud, zero-config. cce init auto-detects your editor.

---

Install and see savings in 60 seconds

uv tool install code-context-engine   # or: pipx install code-context-engine
cd /path/to/your/project
cce init                              # index, install hooks, register MCP server

Restart your editor. Done. Every question now hits the index instead of re-reading files.

cce init auto-detects your editor and writes the right config:

Editor	Config written	Instructions
Claude Code	.mcp.json	CLAUDE.md
VS Code / Copilot	.vscode/mcp.json
Cursor	.cursor/mcp.json	.cursorrules
Gemini CLI	.gemini/settings.json	GEMINI.md
OpenAI Codex	.codex/config.toml

Multiple editors in the same project? All get configured in one command.

  my-project · 38 queries

  ⛁ ⛁ ⛁ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶  93% tokens saved

  Without CCE   48.0k  tokens   $0.24
  With CCE       3.4k  tokens   $0.02
  ──────────────────────────────────────────
  Saved         44.6k  tokens   $0.22

  Cost estimate based on Opus input pricing ($5/1M tokens)

---

Why this matters

Input tokens are 85-95% of your Claude Code bill. CCE cuts them by 93% (benchmarked on FastAPI).

Without CCE:    Claude reads payments.py + shipping.py   = 45,000 tokens
With CCE:       context_search "payment flow"            =    800 tokens

	Without CCE	With CCE
Session startup	Re-reads files every time	Queries the index
Finding a function	Read entire 800-line file	Get the 40-line function
Cross-session memory	None	Decisions + code areas persisted
Token cost (Opus, medium project)	~$0.48/session	~$0.14/session

---

Benchmark: FastAPI (independently verified)

We benchmarked CCE against FastAPI (48 source files, 19K lines of Python) with 20 real coding questions. No cherry-picking, no synthetic queries.

Methodology: For each query, "without CCE" means reading the full content of every file the query touches. "With CCE" means the relevant chunks after compression. This is conservative (agents often read more files than needed).

Metric	Result
Retrieval	93% savings (75,355 → 5,381 tokens/query)
+ Compression	90% additional (5,381 → 541 tokens/query)
Combined	99.3% (75,355 → 541 tokens/query)
Recall@10 (found the right files)	0.80
Precision@10	0.30
Latency p50	0.4ms
Queries tested	20

Per-Layer Savings (each measured independently)

Layer	What it does	Savings	Method
Retrieval	Full files → relevant code chunks	93%	measured
Chunk Compression	Raw chunks → signatures + docstrings	90%	measured
Output Compression	Reduces Claude's reply length	65%	estimated
Grammar	Drops articles/fillers from memory text	13%	measured

Reproduce it yourself:

pip install code-context-engine
python benchmarks/run_benchmark.py --repo https://github.com/fastapi/fastapi.git --source-dir fastapi

Full results in benchmarks/results/fastapi.md. Queries and methodology in benchmarks/.

---

What you get

9 MCP tools that Claude uses automatically:

Tool	What it does
context_search	Hybrid vector + BM25 search with graph expansion
expand_chunk	Full source for a compressed result
related_context	Find code via graph edges (calls, imports)
session_recall	Recall decisions from past sessions
record_decision	Save a decision for future sessions
record_code_area	Record which files were worked in
index_status	Check index freshness
reindex	Re-index a file or the full project
set_output_compression	Adjust response verbosity (off / lite / standard / max)

Live dashboard with donut charts, file health, and session history:

cce dashboard

Dollar estimates fetched from live Anthropic pricing:

cce savings --all    # see savings across all projects

---

How it works (the short version)

1. Index: Tree-sitter parses your code into semantic chunks (functions, classes, modules). Stored as vector embeddings locally.
2. Search: Claude calls context_search. Hybrid vector + BM25 retrieval finds the right chunks. Code graph adds related files automatically.
3. Compress: Chunks are truncated to signatures + docstrings (or LLM-summarized if Ollama is running).
4. Remember: Decisions and code areas persist across sessions via session_recall.
5. Track: Every query is logged. cce savings shows exactly how much you saved.

Re-indexing after edits takes under 1 second (96% embedding cache hit rate). Git hooks keep the index current automatically.

---

What makes CCE different

It saves where the money is

Output compression tools (like Caveman) save 20-75% on output tokens. Output is 5-15% of your bill. Net savings: ~11%.

CCE saves on input tokens (93% retrieval + 90% compression on FastAPI, independently benchmarked). Input is 85-95% of your bill.

It actually understands your code

Not a text search. Tree-sitter AST parsing creates semantic chunks. Hybrid retrieval merges vector similarity with BM25 keyword matching via Reciprocal Rank Fusion. A confidence scorer blends similarity (50%), keyword match (30%), and recency (20%). Graph expansion walks CALLS/IMPORTS edges to pull in related code.

It remembers

record_decision("use JWT for auth", reason="session tokens flagged by legal") is stored in SQLite and surfaces via session_recall in the next session. No re-explaining your architecture.

It tracks real savings

Not estimates. Actual tokens served vs full-file baseline, broken down by buckets (retrieval, compression, output, memory, grammar). Dollar costs fetched from Anthropic's pricing page. Savings summary shown at every session start.

It is secure by default

Secret files (.env, *.pem, credentials.json) are never indexed. Content is scanned for AWS keys, GitHub tokens, Slack tokens, Stripe keys, JWTs, and generic credentials. PII (emails, IPs, SSNs, credit cards) is scrubbed from memory writes. All MCP file paths are validated against path traversal.

---

Under the hood

Content-Hash Embedding Cache

SHA-256 fingerprint per chunk, salted with model name. Re-index skips unchanged code. Binary float32 storage (10x smaller than JSON). Typical re-index: 96% cache hit, under 1 second.

sqlite-vec: 2 MB instead of 217 MB

Replaced LanceDB with sqlite-vec. Same cosine-distance quality, 99% smaller install. WAL mode + PRAGMA NORMAL for 80% write speedup. Vectors, FTS5, code graph, and compression cache all in three SQLite files.

Deterministic Grammar Compression

Memory entries compressed without LLM calls. Drops articles, fillers, pronouns. Three levels (lite/full/ultra, 20-60% savings). Code, paths, URLs preserved byte-for-byte. Same input always yields same output.

Fail-Closed Hook Design

5 Claude Code lifecycle hooks capture session context. Every hook runs curl ... || true, so a crashed server never blocks the user. SessionStart injects bootstrap context; others capture silently.

Dynamic Pricing

Dollar estimates in cce savings come from live Anthropic pricing (HTML table parsed, cached 7 days, offline fallback). No manual updates when rates change.

Append-Only Savings Ledger

7 buckets track every token saved: retrieval, chunk compression, output compression, memory recall, grammar, turn summarization, progressive disclosure. Survives restarts. Powers CLI and dashboard analytics.

---

CLI at a glance

cce init                    # Index + install hooks + register MCP
cce                         # Status banner
cce savings                 # Token savings with dollar estimates
cce savings --all           # All projects
cce dashboard               # Web dashboard with live charts
cce search "auth flow"      # Test a query
cce status                  # Index health + config
cce services                # Ollama + dashboard + MCP status
cce commands add-rule '...' # Project rules for Claude
cce uninstall               # Clean removal of all CCE artifacts

Run cce list for the full command reference.

---

Configuration

Zero-config by default. Override what you need in ~/.cce/config.yaml or .context-engine.yaml:

compression:
  level: standard          # minimal | standard | full
  output: standard         # off | lite | standard | max

retrieval:
  top_k: 20
  confidence_threshold: 0.5

pricing:
  model: opus              # opus | sonnet | haiku

---

Output Compression

CCE also compresses Claude's responses (same concept as Caveman):

Level	Style	Savings
off	Full output	0%
lite	No filler or hedging	~30%
standard	Fragments, drop articles	~65%
max	Telegraphic	~75%

Tell Claude: "switch to max compression" or "turn off compression". Code blocks and commands are never compressed.

---

Disk Footprint

Component	Size
Installed package	~189 MB (ONNX Runtime is 66 MB of that)
Embedding model (one-time download)	~60 MB
Index per project (small/medium/large)	5-60 MB

No GPU required. Embedding model runs on CPU via ONNX Runtime.

---

Supported Languages

AST-aware chunking (10 extensions):

Language	Extensions
Python	.py
JavaScript	.js, .jsx
TypeScript	.ts, .tsx
PHP	.php
Go	.go
Rust	.rs
Java	.java

Fallback chunking: All other text files (Markdown, YAML, config, etc.) chunked by line range.

---

Documentation

Page	Content
Examples	Real conversations with Claude
How It Works	Full 9-stage pipeline
CLI Reference	Every command with output
Configuration	All config options
Project Commands	Rules and preferences for Claude
Tech Stack	Every library and why

---

Roadmap

• Semantic indexing + hybrid retrieval + graph expansion
• Cross-session memory (decisions, code areas, session recall)
• Web dashboard with live charts
• Token savings tracking with dollar estimates
• Output compression (off / lite / standard / max)
• Content-hash embedding cache (96% hit rate on re-index)
• sqlite-vec migration (99% smaller install)
• Dynamic pricing from Anthropic docs
• 7-layer security (secrets, PII, path traversal, audit log)
• Clean uninstall (removes all CCE artifacts)
• AST-aware chunking for PHP, Go, Rust, Java (tree-sitter)
• Multi-editor support (Cursor, VS Code/Copilot, Gemini CLI)
• Reproducible benchmark suite (93% savings on FastAPI, per-layer breakdown)
• Session savings visibility (shown at every session start)
• Tree-sitter support for C, C++, Ruby, Swift, Kotlin
• Docker support for remote mode

---

Contributing

Contributions welcome. See https://github.com/elara-labs/code-context-engine/blob/main/CONTRIBUTING.md for setup.

---

License

MIT. See LICENSE.

Authors

• Fazle Elahee
• Raj

Acknowledgments

Claude Code · MCP · sqlite-vec · Tree-sitter · fastembed · Ollama

---

If CCE saves you tokens, give it a star.