Vibe Coding Discoverentroly
View on GitHubCut AI context cost without trusting the compressor. Every reduction is reversible, byte-exact recoverable, and carries an auditable receipt. Local-first, works through proxy, MCP, SDK, or agent wrapper.
Local-first Context Assurance layer that selects budgeted, query-relevant context for AI agents and compresses it reversibly. Every reduction ships an auditable receipt plus a handle to recover the exact original bytes. Integrates via MCP, proxy, SDK, or agent wrapper.
Use Cases
Cut token usage and cost for AI coding agentsSelect only query-relevant code context before sending to an LLMRecover exactly the omitted original bytes via receiptsAudit what context was kept or dropped and whyRun as an MCP server for agent tool callsIntercept LLM API traffic through a local proxyIntegrate context compression via SDK middlewareDetect hallucinations via provenance tracing
Built With
- Language
- Python
- Frameworks
- MCP · LangChain · LiteLLM · CrewAI · AutoGen · Vercel AI SDK · OpenAI SDK · Anthropic SDK · Agno · Strands Agents · Claude Code · VS Code Copilot · OpenCode · OpenClaw · Aider
Tags
context-compression · token-optimization · context-engineering · mcp · ai-agents · llm-proxy · receipts · reversible-compression · local-first · claude-code · github-copilot · cursor · codex · hallucination-detection · developer-tools · llm-cost-optimization