code-context-engine

uvx --from "code-context-engine[local]" cce init    # install + index + configure, one shot

uv tool install "code-context-engine[local]"    # or: pipx install "code-context-engine[local]"
cd /path/to/your/project
cce init

my-project · 38 queries · last query 5m ago

  ⛁ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶  88% tokens saved

  Input savings   1.9M  tokens   $27.78
  Output savings  4.8k  tokens   $0.36
  ──────────────────────────────────────────
  Total saved   1.9M  tokens   $28.15

  Breakdown:
    retrieval              84%  ▰▰▰▰▰▰▰▰▰▰    1.8M   $26.76 · 12 calls
    chunk compression       3%  ▰▱▱▱▱▱▱▱▱▱   68.5k    $1.03 · 12 calls
    output compression*    
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.

Multi-Provider Pricing

Dollar estimates in `cce savings` support 15+ models across Anthropic, OpenAI, and Google. Static pricing ships with CCE, live Anthropic pricing is fetched and cached 7 days. Configure `pricing.model` (e.g. `gpt-4o`, `gemini-2.5-pro`, `sonnet`) or override with `pricing.input` / `pricing.output` for custom rates.

Resource Governor (Multi-Instance Safety)

Running dozens of `cce serve` processes (one per project per AI session) can exhaust system memory. The resource governor caps ONNX Runtime threads per process, uses advisory file locks so only one process indexes a given project at a time, backs off under Linux memory pressure (PSI), and auto-shuts down idle servers after 30 minutes. Configure via `serve.idle_timeout_minutes` and `serve.max_ort_threads`.

Memory Nudges

CCE's cross-session memory depends on the agent calling `record_decision` and `record_code_area`. Memory nudges make recording ambient: after N searches without a recording, `context_search` results include a short reminder. At session end, the Stop hook summarizes unrecorded activity. Nudges re-arm after the first recording so they stay useful without being noisy.

HTTP Search Endpoint

`cce serve --http` exposes a `POST /search` endpoint for custom agent integrations that speak HTTP instead of MCP stdio. Same hybrid retrieval pipeline, structured JSON response with confidence scores. Input validation clamps `top_k` (1..100) and `confidence_threshold` (0.0..1.0).

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.

---

## Agent Plugin

[Agent Plugins](https://agent-plugins.org) is an open standard (v1.0.0) backed by Amazon, Cursor, Microsoft, OpenAI, and Vercel for packaging AI skills and MCP servers into portable, zero-install bundles. CCE can generate a plugin directory that compatible editors can discover and load automatically.

### What gets generated

### Compatible editors

VS Code, GitHub Copilot, ChatGPT, Codex, Cursor, and Kiro. The plugin uses `uvx` to launch CCE on demand, so users do not need to pre-install the Python package. The MCP server auto-discovers the project root by walking up from its working directory, looking for `.context-engine.yaml` or `.git/`.

### When to use `--plugin` vs `--agent`

| | `--agent` (default) | `--plugin` |
|---|---|---|
| Install method | Writes editor-specific config files | Generates a portable plugin directory |
| Zero-install | No, CCE must be on PATH | Yes, `uvx` fetches CCE on demand |
| Instruction updates | Stale until `cce init` re-run | Stale until `cce init --plugin` re-run |
| Best for | Your own machine | Sharing with a team or distributing |

Both can be used together. `--agent` handles per-editor MCP config, `--plugin` provides a portable alternative.

---

## CLI at a glance

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`:

**Remote Ollama:** If you run Ollama on another machine in your network, set `compression.ollama_url` (e.g. `http://nas.local:11434`) or export `CCE_OLLAMA_URL` (the env var wins). CCE probes the endpoint and falls back to truncation-only compression when it's unreachable, so a flaky link won't break indexing.

---

## 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 |
|-----------|------|
| Core install (Ollama backend) | ~17 MB |
| With `[local]` extra (fastembed + ONNX) | ~189 MB |
| Embedding model (one-time download) | ~60 MB (fastembed) or managed by Ollama |
| Index per project (small/medium/large) | 5-60 MB |

No GPU required. With Ollama, embeddings are handled by the Ollama server. With the `[local]` extra, the embedding model runs on CPU via ONNX Runtime.

---

## Supported Languages

**AST-aware chunking (tree-sitter parsed, 11 extensions):**

| Language | Extensions |
|----------|-----------|
| Python | `.py` |
| JavaScript | `.js`, `.jsx` |
| TypeScript | `.ts`, `.tsx` |
| PHP | `.php` |
| Go | `.go` |
| Rust | `.rs` |
| Java | `.java` |
| C# | `.cs` |

**Language-aware fallback chunking (40+ extensions):**

| Category | Languages |
|----------|-----------|
| Web | HTML, CSS, SCSS, LESS, Vue, Svelte |
| Systems | C, C++, Zig, Nim |
| Mobile | Swift, Kotlin, Dart |
| Functional | Haskell, Scala, Clojure, Elixir, Erlang, F# |
| Scripting | Ruby, Perl, Lua, R, Bash/Zsh |
| Data/Config | JSON, YAML, TOML, XML, SQL, GraphQL, Protobuf |
| DevOps | Terraform, HCL, Dockerfile |
| Docs | Markdown |

All other text files are chunked by line range. Binary files are skipped.

---

## Documentation

| Page | Content |
|------|---------|
| [How Much Are You Spending on AI Coding Tokens?](https://elara-labs.github.io/code-context-engine/blog/real-cost-of-ai-coding-tokens.html) | The math on input vs output tokens |
| [What is CCE? (Complete Guide)](https://elara-labs.github.io/code-context-engine/blog/what-is-code-context-engine.html) | Setup, tools, how it works, FAQ |
| [How to Save Claude Code Tokens](https://elara-labs.github.io/code-context-engine/blog/save-claude-code-tokens.html) | Cost breakdown and savings guide |
| [Benchmark Deep Dive](https://elara-labs.github.io/code-context-engine/blog/benchmark-fastapi.html) | Full FastAPI benchmark methodology |
| [Comparison with Alternatives](https://elara-labs.github.io/code-context-engine/comparison.html) | CCE vs Cursor, Aider, Continue, Greptile |
| [Examples](https://github.com/elara-labs/code-context-engine/blob/main/docs/wiki/Examples.md) | Real conversations with Claude |
| [How It Works](https://github.com/elara-labs/code-context-engine/blob/main/docs/wiki/How-It-Works.md) | Full 9-stage pipeline |
| [CLI Reference](https://github.com/elara-labs/code-context-engine/blob/main/docs/wiki/CLI-Reference.md) | Every command with output |
| [Configuration](https://github.com/elara-labs/code-context-engine/blob/main/docs/wiki/Configuration.md) | All config options |

---

## FAQ

### Does CCE affect response quality?

No. Quality stays the same or slightly improves.

CCE replaces "dump the entire file" with "search for the relevant function." The model still gets the code it needs (0.90 Recall@10 in benchmarks). Less irrelevant context means less noise competing for attention, which can improve the model's focus on your actual question.

### How does output token savings work?

CCE writes output compression rules directly into your agent's instruction files (`CLAUDE.md`, `AGENTS.md`, `.cursorrules`, etc.) during `cce init`. These rules apply to the **entire session**, not just CCE tool responses, so every reply from the agent follows them.

Set the level in `~/.cce/config.yaml` or `.context-engine.yaml`:

Then re-run `cce init` to update instruction files. Or change at runtime:

| Level | Savings | What it does |
|-------|---------|--------------|
| `off` | 0% | No compression |
| `lite` | ~25% | Removes filler/hedging/pleasantries + diff-only for code changes |
| `standard` | ~70% | Drops articles, fragments, short synonyms + diff-only for code |
| `max` | ~80% | Telegraphic style + diff-only for code |

Default is `standard`. All levels include **code output rules** that tell the model to show only changed lines (not full file rewrites), which is where most output tokens go in coding sessions. The `max` level produces very terse prose (similar to "caveman mode"). Code blocks, paths, and commands are never compressed regardless of level.

### Where do the savings come from?

Most savings are **input tokens** (what goes into the model):

| Layer | Type | Typical savings |
|-------|------|-----------------|
| Retrieval | Input | 94% (full files → relevant chunks) |
| Chunk compression | Input | 89% (chunks → signatures) |
| Grammar compression | Input | 13% (article/filler removal) |
| Turn summarization | Input | varies (session history) |
| Progressive disclosure | Input | varies (tool payloads) |
| Output compression | Output | 25-80% (depends on level) |

Output tokens cost 5x more per token (e.g. Opus: $15/1M input vs $75/1M output), so even a small output reduction has outsized cost impact.

---

## Roadmap

- [x] Multi-repo benchmarks (FastAPI, chi, fiber)
- [x] More benchmarks (Django, Express)
- [ ] Tree-sitter support for C, C++, Ruby, Swift, Kotlin
- [ ] Docker support for remote mode
- [ ] Port to mcp 2.x API

See [CHANGELOG.md](CHANGELOG.md) for shipped features.

---

## Contributing

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

---

## License

MIT. See [LICENSE](LICENSE).

## Authors

- [Fazle Elahee](https://github.com/fazleelahhee)
- [Raj](https://github.com/rajkumarsakthivel)

## Acknowledgments

[Claude Code](https://docs.anthropic.com/en/docs/claude-code) · [MCP](https://modelcontextprotocol.io) · [sqlite-vec](https://github.com/asg017/sqlite-vec) · [Tree-sitter](https://tree-sitter.github.io/) · [fastembed](https://github.com/qdrant/fastembed) · [Ollama](https://ollama.com/)

---

  If CCE saves you tokens, give it a star.