Open and maintain a live Vibe Canvas projection beside the current Codex conversation. Use when the user explicitly asks to open or start Vibe Canvas, keep syncing later turns without another submit action, let AI auto-Pick or buffer turns, and preserve manual Pick as the final authority.
Delegate implementation to an external coding agent through Agent Bridge. Use when Codex should supervise Claude Code or DeepSeek Harness instead of editing the target repo itself. Covers bridgerun, continue, respond, approve, apply, reject, cancel, doctor, agents, prune.
Checks that the citations and evidence behind a hypothesis, review or write-up actually say what they are claimed to say, and reports the ones that do not. Use when the user asks to verify sources, check citations, confirm a claim is supported, or before publishing, sharing or acting on anything the co-scientist…
Playbook for using the Curtis server — preparing the contact list, writing the messages, reading the right numbers, and repairing selectors when LinkedIn changes the DOM. Use it when the user wants to send connection requests or messages on LinkedIn with this tool.
Use Codex as the only natural-language Agent for warehouse/logistics CAD. Codex turns free Chinese into a strict, inspectable CadProgram, asks for missing drawing facts, then invokes deterministic engine operations for source audit, blank structure copies, editable 2D, same-camera comparison, confirmed height-driven…
★not rated 0
changed 3d agoA83 tokens
originalApache-2.0
Supervise a coding task through the dsh-gate MCP tools, including durable reconnect, interaction handling, observation cursors, the five-minute aggregated progress cadence, and strict completion validation. Use when the user asks Codex to delegate or supervise work in a DSH Host. Do not use for ordinary local coding…
Inspect and review a live native macOS AppKit interface in Codex Browser. Use when the user asks Codex to preview a running AppKit app, click or identify an NSView, map a visual element to its class and hierarchy, diagnose layout or accessibility issues, or prepare precise visual feedback for Codex.
Discover approved Git and non-Git projects and safely inspect, edit, test, and use guarded Git delivery when available through ChatGPT Workspace MCP tools.
Start and use the installed Rhino MCP bridge for an explicitly routed Rhino or Grasshopper request. Use when the user invokes $rhino-mcp or begins the request with /RhinoMCP; do not use for unrelated work.
Use installed Wallpaper Engine projects as a reversible background skin inside the Codex app. Use when the user asks to skin, theme, decorate, or set a Wallpaper Engine background in Codex, list compatible local projects, remove the Codex skin, or generate a native Codex color theme.
Read bounded UTF-8 files and create new files or directories on explicitly configured Windows folders through PC FileBridge. Use when the user asks ChatGPT or Codex to inspect, find, or create local PC files without overwriting or deleting anything.
Launch or reuse the loopback Memex Workspace UI when the user asks to browse conversations, manage facts, inspect taxonomy or pipeline health, or explore the knowledge map.
Synthesize per-content visual strategies from project, audience, content, assets and confirmed learning; then route six asset channels, produce candidates or formal sets, and enforce Renderer text, QA, G4, Style Lock and explicit rule governance.
Coordinate a bounded coding loop where Luna/Codex executes locally and, with explicit browser-control consent, asks ChatGPT Sol to plan or review through a workspace-safe Hua PlanRelay connection. Use when the user asks to run, plan, or review work with Hua PlanRelay; do not use for ordinary coding that does not…
Recall and maintain one Agent's private, sourced experience together with human-reviewed shared ART Knowledge Editions. Use when prior decisions, procedures, project facts, user preferences, earlier failures, or reusable conclusions may affect the current task, and when the user asks to remember, retrieve, correct, or…
Run long-lived autonomous agent loops with persistent memory, context-window compaction, self-healing retries, and resumable state. Use when an AI coding agent must retain progress across sessions, recover from failures, or avoid context exhaustion.
Query the APITube News API — authenticate, search and filter worldwide news articles by keyword, entity, sentiment, source, country and date, and parse the JSON response. Use when an agent needs live or historical news data, media monitoring, or news for a RAG pipeline.
Turn outcome-level requests into a durable project contract, capability-aware execution graph, verified result, and resumable handoff. Use for projects, products, research, automation, migrations, or other multi-step work that must reach an observable finish; do not use for a single atomic answer or edit.
Establish a short, evidence-backed decision gate before architecture or broad implementation when missing facts, requirements, external behavior, or acceptance criteria could change the direction. Use only for consequential or materially uncertain work.
A system for running a long-term team of coding agents on one or more real project directories. It organizes work as Stories, which are trackable pieces of work with a complete, verifiable outcome.
Use Claude as an independent second-model reviewer of Codex work. Supports normal and adversarial reviews, exact commit scope, working-tree or branch scope, explicit files, background jobs, wait-for-result delivery, status, result, and cancellation.
★not rated 0 2mo agoA52 tokens
originalMIT
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: