Shared local context and memory for multiple AI coding agents via MCP, scoped facts, startup injection, and secret-safe distillation.
This repository also configures its own agents. See what unified-agent-context tells them →
anthonycho-ux/unified-agent-context
Skill Claude CodeCodex
Schmidhuber compression progress as intrinsic curiosity reward for learning systems, curriculum design, and exploration. Use when designing curiosity-driven agents, progress-based curricula, or measuring learning as compression improvement.
anthonycho-ux/unified-agent-context
Skill Claude CodeCodex
Pool and route free model endpoints when a coding agent hits a rate limit or auth failure (429/401/503). Use when the user says "hit rate limit", "key expired", "401/429", "route to a free model", "pool free models", or a provider/key is down and you need a working fallback model.
anthonycho-ux/unified-agent-context
Skill Claude CodeCodex
Write to the user not just like a human, but like one of the Korean essayists they love is really speaking. Kim Hoon voice by default, other voices swapped in by purpose. Use for every conversational reply, summary, or explanation, and especially when the output sounds like a lecture, a report, or like AI wrote it.…
anthonycho-ux/unified-agent-context
Skill Claude CodeCodex
Publish posts on Threads (threads.com) for the user. Use whenever the user wants to write, draft, or publish a Threads post or thread, with or without an attached image, including "post this", "write a post using this image", or a pasted block of raw post text.
anthonycho-ux/unified-agent-context
Skill Claude CodeCodex
Render designed dark-theme card images (diagrams, timelines, comparisons, contracts) from small HTML fragments for image-first communication with the user. Use whenever a reply should include a visual card. Token-efficient - the CSS design system is paid once in the template; each card costs only its content fragment.
anthonycho-ux/unified-agent-context
Skill Claude CodeCodex
Author web animations with Anime.js v4 using the exact v4 API surface: animate(), createTimeline(), createTimer(), createAnimatable(), createDraggable(), createLayout(), createScope(), waapi, engine, easings, utils, svg and text modules. Use when writing or debugging Anime.js animations, timelines, staggered effects…
anthonycho-ux/unified-agent-context
Skill Claude CodeCodex
Design long-form reading pages with editorial minimal, a content-first page design system for single-page essay sites, editorial layout, and long-form web story pages where the writing is the product. One signature element carries the identity, everything else stays quiet.
anthonycho-ux/unified-agent-context
Skill Claude CodeCodex
Converse with coding agents (Claude Code, Codex, Grok, ...) that live in Herdr panes from another Herdr-managed pane - resolve the agent behind a pane, submit a prompt, wait for the reply, read the response, and relay it to the user. Use when the user says "ask claude", "check with the other agent", "have codex look…
anthonycho-ux/unified-agent-context
Skill Claude CodeCodex
Audience: orchestrator agents, primarily Claude Desktop. Kimi Code CLI ("kimi") is a full-stack execution agent running on the user's Mac. It can be invoked non-interactively via one-shot kimi -p calls. Verified 2026-07-24.
anthonycho-ux/unified-agent-context
Skill Claude CodeCodex
Join and stay reactive in the shared multi-agent meeting room (an append-only coordination file on neutral ground). Use whenever the user or another agent points you to a meeting room path, asks you to pay attention there, or when coordinating several agents across panes or hosts without a central server. Covers the…
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: