Context engineering

Extend Pydantic AI agents with batteries-included capabilities from pydantic-ai-harness -- Code Mode (collapse many tool calls into one sandboxed Python execution), a filesystem and shell…

pydantic/skills

Audit and remediate Entroly's MCP marketplace quality with evidence, adversarial validation, and no score gaming.

juyterman1000/entroly

Starts a new Claude Code session from only the last quarter of the current conversation, dropping earlier context to save tokens while keeping recent work.

ykdojo/claude-code-tips

Build MCP servers for customer care agents following Watson Orchestrate specifications.

IBM/ibm-watsonx-orchestrate-adk

Audit and improve repositories for reliable agentic work across Codex and Codex App, Claude Code, and OpenCode.

CodeAlive-AI/ai-driven-development

Extend context windows of transformer models using RoPE, YaRN, ALiBi, and position interpolation techniques.

Orchestra-Research/AI-Research-SKILLs

Asks adaptive questions about a project phase and records the decisions in a CONTEXT.md that later research and planning agents can act on without asking again.

open-gsd/gsd-core

Run an autonomous, spec-driven development "saga" for medium-to-large features using an orchestrator agent and a fleet of worker subagents.

warpdotdev/common-skills

GPUI framework knowledge covering actions/keybindings, async/background tasks, context management (App/Window/Context<T/AsyncApp), custom elements (low-level Element trait), entity state management…

feigeCode/navop

Validates the context file for a Nucleus V2 workflow run, discovers standalone context from explicit input, and writes a blocked result when required context is missing.

protect-my-hair/nucleus-marketplace

Catalog of obstacles, anti-patterns and patterns for working with AI coding agents, covering context management and reliability, from a published patterns collection.

lexler/skill-factory

Configures, repairs and audits a Codex setup that uses a one-million-token context through a direct OpenAI API route while keeping the ChatGPT login for connectors.

steipete/agent-scripts

Label each context chunk keep, drop or replace-with-a-pointer and pass the survivors through byte for byte instead of summarising, with key-shaped chunks decided locally and never sent, and a…

mrmps/classifier-dev

A read-only explorer for cold-start orientation or failed searches that replies with nothing but file path and line range citations, keeping its reads out of main context.

JuliusBrussee/caveman

在多轮对话和多 Agent 协作场景下，帮助模型管理指令、项目状态和长上下文。通过外部文档与会话笔记实现可控的“记忆”、指令冲突检测和高质量交接。适用于任务跨多次调用、跨 Agent、需要稳定行为规范时使用。

No-Trade-No-Life/Yuan

Diagnoses which of three failure modes a stalled AI-assisted build is in, then chooses a targeted fix, a hard reset or a rethink instead of more prompting.

jsmastery-pro/jsm-agent-skill

A skill your agent uses when the user says 'tokenstack', 'compression', 'token savings', 'proxy status', or asks about context window usage.

cwinvestments/memstack

Aggressive token, context, and cost optimization mode for AI coding agents (Claude Code, Antigravity, Cursor, etc.).

kipperacademy/skillpper

Persist agent state across runs, shape what the LLM sees per turn, and cap history to fit a context window.

ag2ai/build-with-ag2

Show how much context window context-mode saved this session.

mksglu/context-mode

Spanish edition of a planning skill that keeps a multi-step agent task on track with task_plan.md, findings.md and progress.md on disk, with recovery after a session reset.

OthmanAdi/planning-with-files

Analyze Claude Code session bloat — shows token count, context usage %, and bloat breakdown.

Ruya-AI/cozempic

Orchestrate parallel Claude Code worker swarms with protocol-based behavioral governance.

cj-vana/claude-swarm

Writes a handoff note so a fresh agent can pick up your work without the session history, or finds and resumes from any continuity source you choose.

EveryInc/compound-engineering-plugin

Teaches how xc-plugin saves tokens with progressive disclosure, cached responses and consistent configuration, so large device lists and build logs arrive as summaries first.

conorluddy/xclaude-plugin

Onboard a fresh or early scaffold after Blueprint is overlaid by tuning commands, standards, adapters, visibility, and context loading.

aiblueprinthq/ai-blueprint

Ranks a large catalog of installed skills against the current request through the Jev service, and can conclude that no skill applies.

kerpopule/hermes-jev-skills

Single-pass codebase analysis leveraging a 1M-token context window for comprehensive security scanning, architecture review, and dependency auditing.

yonatangross/orchestkit

Maps an unfamiliar repo or subsystem with the ripwire CLI before reading files, climbing an escalation ladder only until you are oriented.

redhat-et/ripwire

Teaches how to work within MiniMax M3's 1M-token context: decide per source what to keep, summarize or drop, plan the loading, and cap raw blocks across iterations.

madebyaris/advance-minimax-m3-cursor-rules

Searches Mem0 memories from earlier Pi sessions in the repository and prints a short block of the relevant ones before work starts, read-only and silent when empty.

mem0ai/mem0

Searches Mem0 memories from earlier OpenCode sessions in the repository and prints a plain-text block of up to 10 relevant ones before a task begins.

mem0ai/mem0

Keeps plans, decisions, evidence and reviewer verdicts in small files so long multi-phase tasks survive context compaction and session resumes.

ZaxbyHub/opencode-swarm

Apply when context is filling up: large outputs, long files, repeated reads, fan-out planning.

cursor/plugins

Guide for setting up and using the AgenticX memory system including Mem0 integration, long-term memory, context management, and memory-enhanced agents.

DemonDamon/AgenticX

Maintain repository agent guidance. An agent skill from videojs/video.js.

videojs/video.js

Compresses a TypeScript or Swift repository into one symbol-map text file sized for an LLM context window, for refactor planning and architecture recon.

steipete/agent-scripts

Revalidate persistent agent knowledge by classifying why it can become stale and how quickly it changes, then retain, revise, externalize, or retire it.

pskoett/pskoett-ai-skills

Configures OmniRoute's RTK, Caveman and stacked compression modes, manages language packs and rules, and previews token savings of 60 to 90 percent.

diegosouzapw/OmniRoute

Controls the RTK filter set and context-handling settings in OmniRoute, with endpoints to try compression on sample text and read back retained output.

diegosouzapw/OmniRoute

Gives a fast orientation to OpenRig for an agent that just booted into a seat, covering rigs, topologies, layers and where context and skills come from.

mvschwarz/openrig

Autonomous multi-session campaign agent. An agent skill from SethGammon/Citadel.

SethGammon/Citadel

Generate a HANDOFF.md context transfer document for session continuity.

Szotasz/marveen

Keeps a multi-step agent task on track with task_plan.md, findings.md and progress.md on disk, with session recovery and rules for logging errors and progress.

OthmanAdi/planning-with-files

A skill your agent uses when maintaining context init, checkpoint, recover, index, or summarize behavior.

meamaturinlove221/TuringResearch_plus

Audits Claude Code context window consumption across agents, skills, MCP servers, and rules.

affaan-m/ECC

Configure token budget limits, auto-compact settings, and view current budget status (model-aware — Claude 5 lineup, Opus 5.5 default fallback, full 1M context at standard price)

egorfedorov/claude-context-optimizer

Prunes a bloated Claude Code session with the cozempic CLI, shows the token savings, and spawns a watcher that reopens the pruned session in a new terminal window.

Ruya-AI/cozempic