AgentStack
Sign in Publish
MCP servers Skills Toolkits Why AgentStack Sell Docs
Sign in
MCP verified Apache-2.0 Self-run

Context Engine

mcp-contextenginehq-context-engine · by contextenginehq

Open-source platform for deterministic, token-aware context selection for AI agents and LLMs

↗ Repository
— No reviews yet
0 installs
14 views
0.0% view→install

Install

$ agentstack add mcp-contextenginehq-context-engine

✓ scanned · ✓ verified — works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • ✓ Prompt-injection patterns
  • ✓ Secret / credential exfiltration
  • ✓ Dangerous shell & filesystem operations
  • ✓ Untrusted network calls
  • ✓ Known-malicious package signatures

What it can access

  • ✓ Network access No
  • ✓ Filesystem access No
  • ✓ Shell / process execution No
  • ✓ Environment & secrets No
  • ✓ Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

Are you the author of Context Engine? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

Context Engine

The Context Engine is an open-source platform for deterministic, token-aware context selection. It helps AI agents and LLM applications retrieve exactly the right context within budget constraints.

Architecture

context-engine (you are here)
├── context-core        — Core library: scoring, selection, caching
├── context-cli         — CLI: build, resolve, inspect commands
├── context-mcp-server  — MCP server: JSON-RPC 2.0 stdio transport
├── context-compat      — Compatibility test harness
├── context-specs       — Protocol specifications and ADRs
├── context-docs        — Documentation site
├── context-examples    — Example integrations
├── context-sdk-js      — JavaScript/TypeScript SDK (planned)
└── context-sdk-python  — Python SDK (planned)

Quick Start

# Build the CLI
git clone https://github.com/contextenginehq/context-cli.git
cd context-cli
cargo build --release

# Build a context cache from your documents
./target/release/context build --sources ./docs --cache ./my-cache

# Resolve context for a query within a token budget
./target/release/context resolve --cache ./my-cache --query "deployment" --budget 4000

MCP Integration

The Context Engine exposes its capabilities through the Model Context Protocol, making it compatible with any MCP-aware AI agent.

# Start the MCP server
git clone https://github.com/contextenginehq/context-mcp-server.git
cd context-mcp-server
cargo build --release
./target/release/mcp-context-server --cache-root ./caches

Key Properties

  • Deterministic: Same input always produces identical output, byte-for-byte
  • Token-aware: Respects token budgets for LLM context windows
  • Versioned caches: Backward-compatible cache format with version tracking
  • Frozen contracts: v0 output schemas, error codes, and behaviors are locked

Repositories

| Repository | Description | Status | |-----------|-------------|--------| | context-core | Core scoring and selection library | v0.1.0 | | context-cli | Command-line interface | v0.1.0 | | context-mcp-server | MCP server | v0.1.0 | | context-compat | Compatibility test harness | v0.1.0 | | context-specs | Protocol specifications | v0 | | context-docs | Documentation | Active | | context-examples | Example integrations | Active | | context-sdk-js | JavaScript SDK | Planned | | context-sdk-python | Python SDK | Planned |

License

Apache-2.0

"Context Engine" is a trademark of Context Engine Contributors. The software is open source under the [Apache License 2.0](LICENSE). The trademark is not licensed for use by third parties to market competing products or services without prior written permission.

Source & license

This open-source MCP server is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet — be the first.

Write a review

Versions

  • v0.1.0 Imported from the upstream source.

More in AI & ML