@ldelossa/pi-ide
Pi extension that connects to a running editor (Neovim today, more later) over a local WebSocket. Adds ambient editor context to the agent and routes write/edit tool calls through interactive diffs in the editor.
6.2k listings for pi · 0 reviewed by us
696 results · page 9 of 15
Pi extension that connects to a running editor (Neovim today, more later) over a local WebSocket. Adds ambient editor context to the agent and routes write/edit tool calls through interactive diffs in the editor.
🚀 pi-next — The ultimate next-gen enhancement suite for pi coding agent. Agent orchestration dashboard with real-time TUI, semantic compaction 2.0 that preserves decisions & files, streaming bash protocol with live output, context/tool/cost inspector, li
Role-based persona system for AI agents — memory, knowledge, embedding. CLI, MCP, daemon, Pi extension.
Standalone checkpoint and searchable cold-history compaction for Pi
Tool result processing, distillation, and context management for pi-coding-agent
Long-term memory for Pi — local libSQL database with vector search.
Persistent memory for pi coding agent. Hybrid cloud + local backend (Memra).
Pi extension for Cognee AI memory — supports both direct SDK and MCP bridge modes. Remember, recall, forget, and manage knowledge graphs from your coding agent.
Tool-agnostic glossary management for coding agents
Model Context Protocol (MCP) integration for pi-coding-agent. Connect to any MCP server (stdio, SSE, StreamableHTTP, WebSocket) and use its tools, resources, and prompts natively.
Semantic token compression for pi terminal — filters noisy bash output and retroactively masks stale tool results in conversation history (per JetBrains Dec 2025 + Anthropic tool-result clearing)
/ide for pi — connect Neovim/VS Code, auto-inject editor context into LLM messages
Observability extension for pi: token/cost/cache/tool/context stats in footer + /stats command
Hindsight self-hosted memory extension for Pi
Project archaeology and repo memory tooling — CLI-first project-memory engine
MCP (Model Context Protocol) adapter extension for Pi coding agent
Pi extension that renders LaTeX formulas as terminal images while preserving original LLM context.
hasslefree and self-contained mempalace port for pi-coding-agent with SurrealDB-backed vector search, layered retrieval, and multi-process auto-server runtime
Companion extension for pi-telegram (by llblab) — posts a compact live-updating service message listing tools used by the agent during Telegram-originated turns only.
Project-level session memory for the pi coding agent — a single MEMD.md file per project
Custom compaction for pi — swap the model, template, and trigger for context compaction
Pi extension and skill for Mnemosyne — local-first AI memory
Automatic context management for pi sessions: anchors, pivots, and cross-session recall — with a status line and auto-truncation of stale tool results.
Thin Pi lifecycle bridge to local Hermes/Icarus Memory OS hooks and tools
Pi extension for flat peer-agent orchestration — named, addressable teammates with resumable context (mirrors Claude Code's team-mate model)
Intelligent model routing and prompt compression for Pi/Kiro
Remote GPT compaction, proactive triggering, and compact TUI transcript retention for Pi
Concurrent sub-agents for pi. /sub <task> opens a fork-point selector so you can pick any user message as the sub's inherited context (Enter = latest); /subn <task> spawns from latest without the selector; status line shows main/sub1/sub2; alt+s cycles su
Play chess against the LLM in pi's TUI — full rules, annotated moves, disk persistence, and constant context cost
Pi extension that injects project context into first-turn prompts and compaction continuity messages.
Filter and block models from any provider via declarative allow/block rules. Hide unwanted Copilot models, enforce global allowlists, or block by context window size.
MCP (Model Context Protocol) adapter extension for Pi coding agent — bridges any MCP server into Pi as first-class tools
Hide all tools behind a manifest-aware tool_search — LLM enables tools by name on demand
Compact oversized Pi tool results with transparent previews and spill files.
Hindsight-backed persistent memory extension for pi
Developer toolkit for Pi — modular skills for authoring extensions, skills, prompt templates, themes, packages, and context files. Includes introspection tools for version, docs, changelog, benchmarking, and loaded tools.
Structured memory system for Vera agent (SQLite + FTS5)
Give every Pi session a private MCPorter daemon lane.
Pi extension: resolve @filepath (files & directories) in AGENTS.md, inject into system prompt
Discord read tools for pi — pull channel messages and attachments for agent context with an Effect-powered bot client
Ambient HUD injected into the model's context on every LLM call. Live briefing notes (time, context budget, git, cwd) that refresh as assistant messages complete — kept off the transcript and out of the user's view.
GitNexus knowledge graph integration for pi — enriches searches with call chains, execution flows, and blast radius. Supports stdio and HTTP transports.
Persistent memory for the Pi coding agent, tuned for local LLMs. Curated decisions, preferences, and workflows that survive future sessions, with cheap context-aware injection.
Android CLI tools for pi — Gradle, ADB, Emulator, Logcat, Layout Inspector, and Android Knowledge Base integration.
Trail: session artifacts and fresh-session checkpoints for Pi
Model benchmarking with domain-aware hallucination tracking, per-model leaderboards, and task templates. Track speed, quality, and pass rate across coding, reasoning, and general knowledge.
Pi extension for the @jeffs-brain/memory pipeline. Recall, extract, reflect, consolidate on every pi session via four lifecycle hooks plus 11 memory_* tools.
Zettelkasten-based agent memory system with bidirectional links