Acontext and sample-strands-agents-agentskills
About Acontext
memodb-io/Acontext
Agent Skills as a Memory Layer
Implements automatic skill extraction and progressive-disclosure retrieval without embeddings—using a "Skill Agent" to distill conversation traces into editable Markdown files that agents can fetch via tool calls. Supports Python and TypeScript SDKs, integrates with LangGraph and Claude, and persists skills as version-controllable files with optional self-hosting via Docker.
About sample-strands-agents-agentskills
aws-samples/sample-strands-agents-agentskills
Agent Skills implementation for Strands Agents SDK
Implements the AgentSkills.io standard with a Progressive Disclosure architecture that loads skill metadata (~100 tokens) on startup, then instructions and resources on-demand, reducing token overhead compared to traditional tool-based approaches. Provides three implementation patterns: file-based (LLM direct access), tool-based (structured context injection), and meta-tool (isolated sub-agents), enabling flexible integration with the Strands Agents SDK for specialized domain tasks like document generation and file processing.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work