MCP plugin that saves 98% of your context window. Works with Claude Code, Gemini CLI, VS Code Copilot, OpenCode, and Codex CLI. Sandboxed code execution, FTS5 knowledge base, and intent-driven search.
OpenCode plugin that blocks inefficient tools and guides AI to use context-mode MCP tools instead
Long-memory plugin for OpenCode with context-mode interop
One-command onboarding for context-mode Platform. Detects your OS and installed AI agents, then writes platform.json and configures each agent's MCP server in a single npx call.
Fork of context-mode with pi-crew integration. MCP plugin with 96% context savings, FTS5 knowledge base, and sandboxed code execution.
context-mode integration for oh-my-pi (OMP) — saves up to 98% of context window via sandboxed execution and session continuity
hotmilk — Pi package bundling gentle-pi, context-mode, graphify, subagents, and bundled extension toggles via hotmilk.json
Throw, identify, and decode Solana JavaScript errors
React useContextSelector hook in userland
Codex-native Claude review plugin: agentic Opus 4.7 reviewers, including explicit Opus 4.7 1M long-context mode, with full read-only tool access for high-scrutiny code review, adversarial review, deep multi-agent review, and security-focused review.
This package brings together every error message across all Wallet Standard JavaScript modules.
Tiny helpers for processing regex syntax
Stop losing context to large outputs — MCP server that compresses tool outputs. Works everywhere: Claude Code, Cursor, Windsurf, Copilot, and any MCP-compatible IDE.
The next generation of events handling for javascript! New: abstract away the network!
Sandboxed code execution + FTS5 knowledge base. 4 MCP tools, 2 auto-enforcing hooks.
React components, HOCs and hooks to assist with tracking user activity with React components
AWS X-Ray SDK for Javascript
This is an internal utility, not intended for public usage.
Secure memory & context optimization MCP plugin for Claude Code — drop-in replacement for context-mode with credential isolation, SSRF protection, MemGPT-style persistent memory, and A2A multi-agent broadcast channel
MCP plugin that saves 98% of your context window. Works with Claude Code, Gemini CLI, VS Code Copilot, OpenCode, and Codex CLI. Sandboxed code execution, FTS5 knowledge base, and intent-driven search.
Error management for Codama
View docs [here](https://radix-ui.com/primitives/docs/components/context-menu).
A gorgeous reporter for Mocha.js
Various type declarations used across Appium
Provides a Jekyll tag plugin that displays lots of information from context for websites in development mode.
Provides a Jekyll tag plugin that displays lots of information from context for websites in development mode.
The app provides a command-line interface (CLI) to an Ollama AI model, allowing users to engage in text-based conversations and generate human-like responses. Users can import data from local files or web pages, which are then processed through three different modes: fully importing the content into the conversation context, summarizing the information for concise reference, or storing it in an embedding vector database for later retrieval based on the conversation.
CommandSet is a user interface framework. Its focus is a DSL for defining commands, much like Rake or RSpec. A default readline based terminal interpreter (complete with context sensitive tab completion, and the amenities of readline: history editing, etc) is included. It could very well be adapted to interact with CGI or a GUI - both are planned. CommandSet has a lot of very nice features. First is the domain-specific language for defining commands and sets of commands. Those sets can further be neatly composed into larger interfaces, so that useful or standard commands can be resued. Optional application modes, much like Cisco's IOS, with a little bit more flexibility. Arguments have their own sub-language, that allows them to provide interface hints (like tab completion) as well as input validation. On the output side of things, CommandSet has a very flexible output capturing mechanism, which generates a tree of data as it's generated, even capturing writes to multiple places at once (even from multiple threads) and keeping everything straight. Methods that normally write to stdout are interposed and fed into the tree, so you can hack in existing scripts with minimal adjustment. The final output can be presented to the user in a number of formats, including contextual coloring and indentation, or even progress hashes. XML is also provided, although it needs some work. Templates are on the way. While you're developing your application, you might find the record and playback utilities useful. cmdset-record will start up with your defaults for your command set, and spit out an interaction script. Then you can replay the script against the live set with cmdset-playback. Great for ad hoc testing, usability surveys and general demos.
pikuri-vectordb gives a pikuri-core agent a +vectordb_search+ tool over a local document corpus — agentic search, the agent decides when to retrieve. Ships a swappable backend (a pure-Ruby +Backend::InMemory+ for teaching and a thin +Backend::Chroma+ HTTP client for persistence), a chunker, an embedder wrapper over +RubyLLM.embed+, and an optional +Reranker::LlamaServer+ that speaks +/v1/rerank+ against a cross-encoder model. Text extraction goes through +Pikuri::FileType.read_as_text+ in pikuri-core, which handles plain text / Markdown / PDF; HTML extraction is a deferred follow-up. Hosts wire the feature via +c.add_extension Pikuri::VectorDb::Extension.new(...)+ inside the +Agent.new+ block — same opt-in shape as +pikuri-tasks+ / +pikuri-skills+. The bundled +Pikuri::VectorDb::LIBRARIAN+ persona is the privilege-separated sub-agent counterpart for hosts that want recall to flow through a child rather than the parent's context. Three model endpoints in the full setup — chat (via ruby_llm), an embedder (via +RubyLLM.embed+), and an optional reranker (HTTP +/v1/rerank+). A single +llama-server+ in router mode serves all three by default, loading each cached GGUF on demand; see the gem's README for details.
Diff and patch tables
Diff and patch tables
No description provided.
No description provided.
No description provided.
No description provided.
No description provided.
No description provided.
No description provided.
No description provided.
No description provided.
No description provided.
No description provided.
No description provided.