codegrid/SKILL.md
CodeGrid is a native macOS canvas where multiple coding agents (Claude, Codex, Gemini, Cursor, Grok, shells) run side by side in panes and collaborate via a local agent bus — no tmux, no cloud, no account, no stored API keys. Install this skill when an agent should know how to operate inside a CodeGrid pane, drive the workspace from outside (control socket or codegrid:// deep links), spawn or message sibling agents, or coordinate multi-agent work (delegate, review, pipeline, parallel fan-out, monitor, debate). The differentiator: multiple coding agents collaborating on one canvas, addressable by stable session_id, with a read → message → read protocol built for orchestration.
npx skillsauth add bankrbot/skills codegridInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
CodeGrid runs many coding agents side by side on an infinite 2D canvas. Each
pane is one process (claude, codex, gemini, cursor-agent, grok, or a
shell) with its own working directory, git branch, and stable session_id. The
agent bus lets any pane discover, read, and message any other pane locally.
This skill is the umbrella entry point. Two references cover the full surface:
Operating CodeGrid — the mental model
(canvas / pane / workspace / agent / bus), MCP tools (list_agents,
read_pane, message_agent), the control-socket JSON-RPC API
(agent_list, agent_read, agent_send, open_folder, new_session,
new_workspace), the codegrid:// deep-link scheme, and an operating
playbook with recipes.
Agent-bus collaboration — the
read → message → read protocol, identifying agents by role and
session_id, orchestration patterns (delegate, review, pipeline, parallel
fan-out, monitor, debate), etiquette and scope safety, loop/runaway
prevention, failure recovery, and worked end-to-end examples.
| You're about to… | Load |
|---|---|
| Discover what's running, open a folder, spawn a session, or drive CodeGrid from outside | using-codegrid |
| Hand work to / consult / review with another agent already running in CodeGrid | codegrid-agent-bus |
| Both | Load using-codegrid first, then codegrid-agent-bus |
list_agents() # discover panes (session_id, role, status)
read_pane(<session_id>) # see what they're doing — always safe
message_agent(<session_id>, "[from <you>] <self-contained request>")
… wait, then read_pane(<session_id>) for the reply.
SOCK="$(cat ~/.codegrid/socket-path 2>/dev/null || echo ~/.codegrid/socket)"
printf '%s\n' '{"jsonrpc":"2.0","id":1,"method":"agent_list"}' | nc -U "$SOCK"
# Or deep-link the installed app:
open "codegrid://open?path=/abs/path/to/repo&type=codex"
open "codegrid://new"
install the codegrid skill from https://github.com/BankrBot/skills/tree/main/codegrid
skills/
(skills/using-codegrid/SKILL.md, skills/codegrid-agent-bus/SKILL.md).
This Bankr Skills provider is a published copy kept in sync.0x6B456E66524aEC1792013eF9DFE87e3F84311ba3 —
see https://codegrid.app/token.read_pane is free; never message a busy agent.session_id, never pane number — pane numbers shift.Full guidance: see the two reference files.
data-ai
Discover, bet on, track, and settle Hunch prediction markets in natural language. Trigger when a user wants to bet, take a position, or get odds on a crypto outcome — token market-cap milestones and flips, launchpad races (Bankr vs pump.fun volume / #1-days / launches over a cap), token head-to-head outperformance, mcap strike-ladders, and up/down price rounds. Also trigger on "what can I bet on about $TOKEN", "odds on …", "take YES/NO on …", "show my Hunch bets", "did my market resolve". Settles in USDC on Base via x402 (≤ $10 / bet); every bet returns an on-chain proof.
tools
HSM-backed secret management for AI agents. Store API keys (including Bankr `bk_` keys), passwords, and credentials in an encrypted vault; retrieve them at runtime via MCP without keeping secrets in chat context. Bankr Dynamic Key Vending issues short-lived scoped `bk_usr_` keys from a partner key (`bk_ptr_`) without manual rotation. Policy-based access control, secret rotation, sharing, EVM transaction intents (sign/simulate/broadcast), multi-chain signing keys, treasury multisig proposals, OIDC federation for external service auth, built-in prompt injection detection, and optional Shroud TEE LLM proxy. Use when the agent needs secure credential storage, just-in-time secret access, guarded on-chain signing, or security scanning — not for Bankr trading prompts, portfolio checks, or x402 calls (use the bankr skill instead).
development
Give your Bankr agent its own brain and a wallet-signed line to every other agent — on any framework, with no API key. SIGNA is the keyless agent layer on Base: resolve any identity to a messageable wallet, send and read wallet-signed DMs, invoke capabilities on the network, and run a brain that reasons on decentralized inference and acts through those capabilities. The Bankr wallet is the only credential. Triggers: "message that agent", "DM this wallet/handle", "reach the agent behind @x", "what is the base market", "resolve @handle to a wallet", "ask the network", "let my agent think and report".
development
AI-powered crypto trading agent, wallet API, and LLM gateway via natural language. Use when the user wants to trade crypto, check portfolio balances (with PnL and NFTs), view token prices, search tokens, transfer crypto, manage NFTs, use leverage (Hyperliquid or Avantis), bet on Polymarket, deploy tokens, set up automated trading, sign and submit raw transactions, call or deploy x402 paid API endpoints, browse the web, or access LLM models through the Bankr LLM gateway funded by your Bankr wallet. Supports Base, Ethereum, Polygon, Solana, Unichain, World Chain, Arbitrum, and BNB Chain.