npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2026 – Pkg Stats / Ryan Hefner

@mcp-abap-adt/llm-agent

v12.0.4

Published

Core interfaces, types, and lightweight default implementations for LLM agent orchestration.

Readme

@mcp-abap-adt/llm-agent

Core interfaces, types, and lightweight default implementations for LLM agent orchestration.

This package is the abstraction layer consumed by @mcp-abap-adt/llm-agent-libs (and transitively by @mcp-abap-adt/llm-agent-server) and by downstream applications that want to build their own agent on our interfaces. It ships:

  • All I* interfaces (IRag, IRagEditor, IRagProvider, IRagRegistry, ILlm, IMcpClient, IPipeline, IClientAdapter, ILlmApiAdapter, IToolCache, ILogger, etc.)
  • Shared types (Message, ToolCall, RagMetadata, CallOptions, Result, errors)
  • Lightweight RAG implementations (InMemoryRag, VectorRag, QdrantRag, InMemoryRagProvider, VectorRagProvider, QdrantRagProvider, SimpleRagRegistry, SimpleRagProviderRegistry, edit strategies, id strategies, corrections module, overlay rags, MCP tool factory)
  • Library helpers usable when embedding SmartAgent in your own server:
    • Resilience: CircuitBreaker, CircuitBreakerLlm, CircuitBreakerEmbedder, FallbackRag
    • LLM call policies: NonStreamingLlmCallStrategy, StreamingLlmCallStrategy, FallbackLlmCallStrategy
    • Tool cache: ToolCache, NoopToolCache
    • API adapters (Anthropic Messages / OpenAI Chat Completions ↔ SmartAgent): AnthropicApiAdapter, OpenAiApiAdapter, with NormalizedRequest / ApiRequestContext / ApiSseEvent / AdapterValidationError
    • Client adapters: ClineClientAdapter
    • Tool utilities: normalizeAndValidateExternalTools, normalizeExternalTools, getStreamToolCallName, toToolCallDelta

For the full default agent runtime (SmartAgent assembly, providers wiring), install @mcp-abap-adt/llm-agent-libs. For the HTTP server and CLI binaries, install @mcp-abap-adt/llm-agent-server (binary-only — not importable as a library). If you only need the helpers above and ship your own server, depending on @mcp-abap-adt/llm-agent is sufficient.

See the repo root for design specs, migration notes, and architectural docs.

Migration from 12.0.0

Symbols that briefly appeared only in @mcp-abap-adt/[email protected] are now in their dedicated packages:

| Symbol(s) | Was (12.0.0) | Now (12.0.1) | |---|---|---| | MCPClientWrapper, McpClientAdapter, LazyConnectionStrategy, PeriodicConnectionStrategy, NoopConnectionStrategy, createDefaultMcpClient | @mcp-abap-adt/llm-agent-server | @mcp-abap-adt/llm-agent-mcp | | makeRag, resolveEmbedder, prefetchEmbedderFactories, prefetchRagFactories, resolvePrefetchedEmbedder, resolveRag, builtInEmbedderFactories, related types | @mcp-abap-adt/llm-agent-server | @mcp-abap-adt/llm-agent-rag | | SmartAgentBuilder, SessionManager, HistoryMemory, HistorySummarizer, DefaultPipeline, PipelineExecutor, HealthChecker, ConfigWatcher, FileSystemPluginLoader, skill managers, RetryLlm, RateLimiterLlm, TokenBucketRateLimiter, LlmReranker/NoopReranker, InMemoryMetrics/NoopMetrics, NoopTracer, NoopValidator, LlmAdapter, LlmProviderBridge, makeLlm, makeDefaultLlm, DefaultModelResolver | @mcp-abap-adt/llm-agent-server | @mcp-abap-adt/llm-agent-libs | | Interfaces: IMetrics, ITracer, ISessionManager, IPluginLoader, IReranker, IOutputValidator, IModelResolver, IMcpConnectionStrategy, IPipeline, pipeline DSL types, health DTOs | @mcp-abap-adt/llm-agent-server | @mcp-abap-adt/llm-agent |

makeLlm, makeDefaultLlm, and makeRag are now async (Promise<ILlm> / Promise<IRag>). Direct callers add one await. makeRag auto-prefetches the backends it needs — no manual warm-up required. resolveEmbedder(cfg, options) remains synchronous and is intended for hot-path callers that need sync resolution — call prefetchEmbedderFactories([...]) once at startup before using this sync resolver. Consumers that build SmartAgent only via SmartAgentBuilder are unaffected (the builder's build() is already async).