okx-ai-guide
Skill264 starsChanged 43 days agoApache-2.0View on GitHub
OKX.AI (the Agent economic system) intro & onboarding entry. Use whenever the user asks what OKX.AI is, what it can do, how to use or get started with it, wants an OKX.AI tutorial / quickstart / help, or types the product name in any spelling / spacing / casing / typo variant (OKXAI, okx ai, okx-ai, lowercase okx.ai, mis-typed Chinese like 啥是okxai) — e.g. what is OKX.AI / OKX.AI 是什么 / 怎么用 OKX.AI / OKX.AI 快速开始, and any paraphrase in any language. Detects the runtime platform, introduces the three roles (User / ASP / Evaluator), and routes the user into identity registration. NOT for generic onchain-OS onboarding, nor for direct task operations (publishing / accepting / delivering / disputing a task — a separate task-lifecycle flow).
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 1 is distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 32
Do NOT call onchainos agent create (or any registration / staking CLI) from this skill.
- Very specific: line 38
Run the detection function below and read its single-line output.
- Distinctive: line 105
Not logged in → hand off to the existing wallet-login flow, then resume the check.
≈ “Common” is an estimate: at least 1 in 5 of the 308,232 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
OKX.AI Guide
The OKX.AI onboarding entry. Introduces OKX.AI (the Agent economic system), detects whether the current runtime can run OKX.AI, and routes the user into one of the three identity-registration flows — or, on an incompatible platform, tells them how to get a compatible one.
Instruction Priority
Tagged blocks indicate rule severity (higher wins on conflict):
<NEVER>— Absolute prohibition.<MUST>— Mandatory step.<SHOULD>— Best practice.
Scope & Boundary
This skill owns: OKX.AI intro + platform detection + login & identity detection (new vs returning user) + routing into registration. It does NOT:
- own the Onchain OS welcome banner — that is
okx-how-to-play. - implement registration — delegated to
okx-agent-identity(see §Step 5). - own the wallet-login flow — Step 1 only checks login via
wallet statusand hands off tookx-agentic-wallet's existing login flow when needed; the registration playbooks also run their own preflight.
<NEVER>
Do NOT call `onchainos agent create` (or any registration / staking CLI) from this skill. Registration is always delegated to `okx-agent-identity`. (Read-only `onchainos wallet status` and `onchainos agent get-my-agents` in Step 1 are allowed — they create nothing.)
</NEVER>Step 0 — Platform detection
<MUST>
Run the detection function below and read its single-line output. `compatible` = output is NOT `unknown`.
</MUST>detect_harness() {
if [ "${CLAUDECODE:-}" = "1" ]; then
echo "Claude Code"
elif [ -n "${HERMES_INTERACTIVE:-}" ] || [ -n "${HERMES_SESSION_SOURCE:-}" ] \
|| [ -n "${HERMES_YOLO_MODE:-}" ] || [ -n "${HERMES_QUIET:-}" ]; then
echo "Hermes"
elif [ -n "${OPENCLAW_CLI:-}" ] || [ -n "${OPENCLAW_SHELL:-}" ]; then
echo "OpenClaw"
elif [ -n "${CODEX_THREAD_ID:-}" ] || [ -n "${CODEX_CI:-}" ]; then
echo "Codex"
else
echo "unknown"
fi
}
detect_harness- Output ∈ {
Claude Code,Hermes,OpenClaw,Codex} → compatible → Step 1. - Output =
unknown→ incompatible → Step 3.
Step 1 — Compatible: login + identity detection (routing gate)
Reached only when Step 0 is compatible. This step decides which page to show — by checking login first, identity second. The order is mandatory: agent get-my-agents requires a logged-in session, so never query identity before login is confirmed.
<MUST>
1. **Login check** — run `onchainos wallet status` and read `loggedIn`.
- `loggedIn: false` → user is not logged in. Do **not** query identity. Hand off to the existing wallet-login flow ([`../okx-agentic-wallet/SKILL.md`](../okx-agentic-wallet/SKILL.md) §login): prompt login, and on success resume here (re-run `wallet status`, then do the identity check).
- `loggedIn: true` → continue to the identity check.
2. **Identity check** — run `onchainos agent get-my-agents`. It returns the logged-in user's own OKX.AI agents on XLayer (identified via JWT).
- **Empty** (no agents) → user has no OKX.AI identity → **Step 2** (role selection page).
- **≥1 agent** → user already has an identity → **Step 4** (registered user home).
</MUST>The branch is decided solely by whether agent get-my-agents returns any agent — never show the role page (Step 2) to a user who already has an identity, nor the registered home (Step 4) to a user with none.
Step 2 — Compatible & unregistered: role selection page
Reached from Step 1 when the user is logged in but has no OKX.AI identity. Render the role-selection page (Variant A) and route the 1/2/3 reply per references/unregistered-role-selection.md (contains Step 2 page + Step 5 routing). Load it when this branch is hit.
Step 3 — Incompatible: intro + install guide
Reached from Step 0 when the platform is incompatible (unknown). No login / identity check applies — OKX.AI cannot run here.
Free zone (1–5 sentences): answer the user's OKX.AI question, then segue.
Fixed zone: render Variant B from references/intro.md in the user's language; substitute {install_doc_url}. Do not offer numbered picks; end the turn.
Step 4 — Compatible & registered: user home
Reached from Step 1 when the user is logged in and already has ≥1 OKX.AI identity. Render the registered-user home (Variant C, filled field-exact from the agent get-my-agents result) and handle its menu replies (Step 6: 1 + Agent ID → that Agent's current tasks; 2 → top ASPs; Register a <role> → register a missing role) per references/registered-home.md. Load it when this branch is hit.
Step 5 — Routing after role pick
Handled in references/unregistered-role-selection.md alongside Step 2 (the 1/2/3 reply → wait-state line + registration playbook).
Step 6 — Registered-home menu routing (from Step 4)
Handled in references/registered-home.md — covers 1 + Agent ID → agent task-in-progress with status mapping, 2 → top ASPs by sales via agent search --query '按销量从高到低排序', and "Register a <role> identity" reroutes.
Acceptance Criteria
detect_harnessreturns the right platform for each marker set; everything else →unknown→ incompatible branch (Step 3).- Compatible branch (Step 1) checks login (
wallet status) before identity (agent get-my-agents) — identity is never queried while logged out.- Not logged in → hand off to the existing wallet-login flow, then resume the check.
- Logged in + no identity → role selection page (Step 2); replying
1/2/3renders the right wait-state and loads the right registration playbook (Step 5). - Logged in + ≥1 identity → registered user home (Step 4), filled from the
agent get-my-agentsresult; the home menu (Step 6) routes1+ an Agent ID → that Agent's current tasks viaagent task-in-progress, mapping each task'sstatusto a label (e.g.2submitted = delivered/awaiting acceptance) rather than blanket-labeling everything "in progress" (withcode=3001→ "not your Agent, re-enter"),2→ top ASPs by sales viaagent search --query '按销量从高到低排序'(backend semantic sort-by-sales).
- Incompatible branch (Step 3) shows the three-role intro (no picks) + install heads-up +
{install_doc_url}; ends the turn. OKX.AI 快速开始/OKX.AI quick starttriggers this skill.- Fixed-zone copy renders in the user's language; emojis / numbers / URLs / placeholders stay literal.
- Zero
onchainos agent createcalls in this skill (only read-onlywallet status/agent get-my-agents); zero Rust changes.
In this repository62 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- docs/llms-full.txtllms.txt
- docs/llms.txtllms.txt
- llms-full.txtllms.txt
- llms.txtllms.txt
- public/llms-full.txtllms.txt
- public/llms.txtllms.txt
- anatomySkill
- authenticate-walletSkill
- build-an-agent-skillSkill
- connect-three-ws-mcpSkill
- create-3d-avatarSkill
- create-a-three-ws-agentSkill
- embed-three-ws-avatarSkill
- find-3d-assetsSkill
- fundSkill
- gcp-triageSkill
- generate-3d-modelSkill
- hire-an-agentSkill
- metamask-agent-walletSkill
- metamask-agent-workflowsSkill
- monetize-serviceSkill
- okx-agent-chatSkill
- okx-agentic-walletSkill
- okx-agent-identitySkill
- okx-agent-payments-protocolSkill
- okx-agent-taskSkill
- okx-ai-supportSkill
- okx-audit-logSkill
- okx-dapp-discoverySkill
- okx-defi-investSkill
- okx-defi-portfolioSkill
- okx-dex-bridgeSkill
- okx-dex-marketSkill
- okx-dex-signalSkill
- okx-dex-socialSkill
- okx-dex-strategySkill
- okx-dex-swapSkill
- okx-dex-tokenSkill
- okx-dex-trenchesSkill
- okx-dex-wsSkill
- okx-growth-competitionSkill
- okx-how-to-playSkill
- okx-onchain-gatewaySkill
- okx-securitySkill
- okx-task-watchSkill
- okx-wallet-portfolioSkill
- pay-for-serviceSkill
- query-onchain-dataSkill
- rig-a-modelSkill
- search-for-serviceSkill
- sell-an-agent-skillSkill
- send-usdcSkill
- tradeSkill
- x402Skill
- changelog-and-docsSkill
- deploy-runbookSkill
- engineering-standardsSkill
- maintain-claude-mdSkill
- The first 60 shown.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

