headroom/wiki/index.md

Headroom

The Context Optimization Layer for LLM Applications

Compress everything your AI agent reads. Same answers, fraction of the tokens.

87% Avg Token Reduction

100% Answer Accuracy

6 Compression Algorithms

100+ LLM Providers

---

What It Does

Every tool call, DB query, file read, and RAG retrieval your agent makes is 70-95% boilerplate. Headroom compresses it away before it hits the model. The LLM sees less noise, responds faster, and costs less.

Your Agent / App
    │
    │  tool outputs, logs, DB reads, RAG results, file reads, API responses
    ▼
 Headroom  ← proxy, Python library, or framework integration
    │
    ▼
 LLM Provider  (OpenAI, Anthropic, Google, Bedrock, 100+ via LiteLLM)

Headroom works as a transparent proxy (zero code changes), a Python function (compress()), or a framework integration (LangChain, Agno, Strands, LiteLLM, MCP).

---

Quick Start

=== "Proxy (Zero Code Changes)"

```bash
pip install "headroom-ai[all]"
headroom proxy
```

```bash
# Point any tool at the proxy
ANTHROPIC_BASE_URL=http://localhost:8787 claude
OPENAI_BASE_URL=http://localhost:8787/v1 your-app
```

That's it. Your existing code works unchanged, with 40-90% fewer tokens.

Want an always-on local runtime instead? See [Persistent Installs →](persistent-installs.md).

=== "Python SDK"

```python
from headroom import compress

result = compress(messages, model="claude-sonnet-4-5-20250929")
response = client.messages.create(
    model="claude-sonnet-4-5-20250929",
    messages=result.messages,
)
print(f"Saved {result.tokens_saved} tokens ({result.compression_ratio:.0%})")
```

Works with any Python LLM client. [Full SDK guide →](sdk.md)

=== "Coding Agents"

```bash
headroom wrap claude       # Claude Code
headroom wrap copilot -- --model claude-sonnet-4-20250514
headroom wrap codex        # OpenAI Codex CLI
headroom wrap aider        # Aider
headroom wrap cursor       # Cursor
headroom wrap openclaw     # OpenClaw plugin bootstrap
```

Starts the proxy, points your tool at it, compresses everything automatically.

If you prefer an always-on proxy that `wrap` can reuse or recover, see [Persistent Installs →](persistent-installs.md).

=== "TypeScript SDK"

```typescript
import { compress } from 'headroom-ai';

const result = await compress(messages, { model: 'claude-sonnet-4-5-20250929' });
// Use result.messages with any LLM client
console.log(`Saved ${result.tokensSaved} tokens`);
```

Works with Vercel AI SDK, OpenAI Node SDK, and Anthropic TS SDK. [Full TS guide →](typescript-sdk.md)

=== "LiteLLM Callback"

```python
import litellm
from headroom.integrations.litellm_callback import HeadroomCallback

litellm.callbacks = [HeadroomCallback()]
# All 100+ providers now compressed automatically
```

---

Framework Integrations

LangChain

Wrap any chat model. Supports memory, retrievers, tools, streaming, async.

from headroom.integrations import HeadroomChatModel

llm = HeadroomChatModel(ChatOpenAI(model="gpt-4o"))

LangChain Guide →

Agno

Full agent framework integration with observability hooks.

from headroom.integrations.agno import HeadroomAgnoModel

model = HeadroomAgnoModel(Claude(id="claude-sonnet-4-20250514"))
agent = Agent(model=model)

Agno Guide →

Strands

Model wrapping + tool output hook provider for Strands Agents.

from headroom.integrations.strands import HeadroomStrandsModel

model = HeadroomStrandsModel(wrapped_model=bedrock_model)
agent = Agent(model=model)

Strands Guide →

MCP Tools

Three tools for Claude Code, Cursor, or any MCP client: headroom_compress, headroom_retrieve, headroom_stats.

headroom mcp install && claude

MCP Guide →

TypeScript SDK

compress(), Vercel AI SDK middleware, OpenAI and Anthropic client wrappers.

npm install headroom-ai

TypeScript SDK Guide →

OpenClaw

ContextEngine plugin for OpenClaw agents. Auto-compresses context in assemble().

headroom wrap openclaw

OpenClaw Plugin →

All integration patterns →{ .md-button }

---

How It Works

Headroom runs a two-stage pipeline on every request:

graph LR
    A[Your Prompt] --> B[CacheAligner]
    B --> C[ContentRouter]
    C --> E[LLM Provider]

    C -->|JSON| F[SmartCrusher]
    C -->|Code| G[CodeCompressor]
    C -->|Text| H[Kompress]
    C -->|Logs| I[LogCompressor]

    F --> E
    G --> E
    H --> E
    I --> E

Stage 1: CacheAligner — Stabilizes message prefixes so the provider's KV cache actually hits. Claude offers a 90% read discount on cached prefixes; CacheAligner makes that work.

Stage 2: ContentRouter — Auto-detects content type (JSON, code, logs, search results, diffs, HTML, plain text) and routes each to the optimal compressor:

Content Type	Compressor	How It Works
JSON arrays	SmartCrusher	Statistical analysis: keeps errors, anomalies, boundaries. No hardcoded rules.
Source code	CodeCompressor	AST-aware (tree-sitter). Preserves function signatures, collapses bodies.
Plain text	Kompress	ModernBERT token classification. Removes redundant tokens while preserving meaning.
Build/test logs	LogCompressor	Keeps failures, errors, warnings. Drops passing noise.
Search results	SearchCompressor	Ranks by relevance to user query, keeps top matches.
Git diffs	DiffCompressor	Preserves change hunks, drops unchanged context.
HTML	HTMLExtractor	Strips markup, extracts readable content.

Context management is handled automatically inside the pipeline (live-zone-only compression): Headroom compresses only the newest content blocks (the latest user message and tool results) and never drops messages from history. The system prompt, tool definitions, and older turns — the provider cache hot zone — are left untouched so prompt caching keeps working.

Nothing is lost. Compressed content goes into the CCR store (Compress-Cache-Retrieve). The LLM gets a headroom_retrieve tool and can fetch full originals when it needs more detail.

Full architecture deep dive →

---

Results

100 production log entries. One critical error buried at position 67.

Metric	Baseline	Headroom
Input tokens	10,144	1,260
Correct answers	4/4	4/4

87.6% fewer tokens. Same answer. The FATAL error was automatically preserved — not by keyword matching, but by statistical analysis of field variance.

Real Workloads

Scenario	Before	After	Savings
Code search (100 results)	17,765	1,408	92%
SRE incident debugging	65,694	5,118	92%
Codebase exploration	78,502	41,254	47%
GitHub issue triage	54,174	14,761	73%

Accuracy Benchmarks

Benchmark	Category	N	Accuracy	Compression
GSM8K	Math	100	0.870	0.000 delta
TruthfulQA	Factual	100	0.560	+0.030 delta
SQuAD v2	QA	100	97%	19% reduction
BFCL	Tool/Function	100	97%	32% reduction
CCR Needle	Lossless	50	100%	77% reduction

Full benchmark methodology → | Known limitations →

---

Key Features

### Lossless Compression (CCR) Compresses aggressively, stores originals, gives the LLM a tool to retrieve full details. Nothing is thrown away. [Learn more →](ccr.md)

### Smart Content Detection Auto-detects JSON, code, logs, text, diffs, HTML. Routes each to the best compressor. Zero configuration needed. [Learn more →](compression.md)

### Cache Optimization Stabilizes prefixes so provider KV caches hit. Tracks frozen messages to preserve the 90% read discount. [Learn more →](ccr.md)

### Image Compression 40-90% token reduction via trained ML router. Automatically selects resize/quality tradeoff per image. [Learn more →](image-compression.md)

### Persistent Memory Hierarchical memory (user/session/agent/turn) with SQLite + HNSW backends. Survives across conversations. [Learn more →](memory.md)

### Failure Learning Reads past sessions, finds failed tool calls, correlates with what succeeded, writes learnings to CLAUDE.md. [Learn more →](learn.md)

### Multi-Agent Context Compress what moves between agents. Any framework. ```python ctx = SharedContext() ctx.put("research", big_output) summary = ctx.get("research") # ~80% smaller ``` [Learn more →](shared-context.md)

### Metrics & Observability Prometheus endpoint, per-request logging, cost tracking, budget limits, pipeline timing breakdowns. [Learn more →](metrics.md)

---

Cloud Providers

Works with any LLM provider out of the box:

headroom proxy                                          # Direct Anthropic/OpenAI
headroom proxy --backend bedrock --region us-east-1     # AWS Bedrock
headroom proxy --backend vertex_ai --region us-central1 # Google Vertex AI
headroom proxy --backend azure                          # Azure OpenAI
headroom proxy --backend openrouter                     # OpenRouter (400+ models)

Or via LiteLLM for 100+ providers (Together, Groq, Fireworks, Ollama, vLLM, etc.).

---

Installation

pip install headroom-ai                # Core library (Python)
pip install "headroom-ai[all]"         # Everything (recommended)
npm install headroom-ai                # TypeScript / Node.js
pip install "headroom-ai[proxy]"       # Proxy server + MCP tools
pip install "headroom-ai[ml]"          # ML compression (Kompress, requires torch)
pip install "headroom-ai[langchain]"   # LangChain integration
pip install "headroom-ai[agno]"        # Agno integration
pip install "headroom-ai[evals]"       # Evaluation framework

Requires Python 3.10+.

---

Next Steps

• Quickstart — Running in 5 minutes
• Integration Guide — Every way to add Headroom to your stack
• Architecture — How the pipeline works under the hood
• Benchmarks — Accuracy and latency data
• Limitations — When compression helps and when it doesn't
• Filesystem Contract — Canonical config/workspace env vars and paths

---

Apache 2.0 — Free for commercial use. GitHub | PyPI | Discord