Read

../okx-agentic-wallet/_shared/preflight.md

. If that file does not exist, fall back to

_shared/preflight.md

. Strategy endpoints require an authenticated Agentic Wallet session — confirm login before running any subcommand.

statusLabel

status

strategyType

strategyType

status

strategyType

Buy Dip

Take Profit

Stop Loss

Buy Above

status

Expired

Cancelling

Cancelled

Failed

Trading

Completed

Creating

Active

Suspended

• mix two languages in one label (pick one — never render the English label and a translation side by side),
• expose the underlying enum name (

  BUY_DIP

  ,

  CHASE_HIGH

  ,

  COMPLETED

  , …) to the user,
• expose the underlying CLI flag value (

  buy_dip

  ,

  chase_high

  ,

  completed

  ,

  cancelled

  , …) to the user,
• pass through the CLI's raw

  statusLabel

  verbatim when the user is conversing in a non-English language — translate it.

• CHASE_HIGH renders as

  Buy Above

  in English (not "Chase High").
• SPEEDING_UP (-4) was removed 2026-05-08 — not a valid filter or display value.

604800

Slippage interpretation reminder (no echo from CLI):

--slippage

input is in percent units —

0.05

= 0.05% (wire

"0.0005"

), NOT 5%. To set 5% slippage, pass

--slippage 5

. The skill / agent should convey this to the user at input time; the CLI itself does not echo a confirmation line.

create-limit

--chain-id

ethereum

eth

1

bsc

56

xlayer

196

solana

sol

501

base

8453

arbitrum

arb

42161

137

10

43114

59144

784

195

Strategy orders are only supported on Ethereum / BSC / X Layer / Solana / Base / Arbitrum right now.

<requested chain>

is not supported — pick one of these to continue.

strategyType

(--direction, --trigger-price, current market price)

--type

1. Parse direction (buy / sell) from user intent ("buy" / "ape in" / "snap up" → buy; "sell" / "take profit" / "stop loss" / "exit" → sell). Passed verbatim as

   --direction <buy|sell>

   .
2. Fetch current price — call

   onchainos market price --chain <chain> --address <token>

   , read

   data[0].price

   . For BUY direction query the to-token's current price; for SELL direction query the from-token's current price. The agent needs this for (a) Step 0 USD-value pre-flight, (b) the Step 1 confirmation page "Trigger Price vs current", and (c) computing the Display label per the table above.
3. Pass

   --current-price <usd>

   to the CLI so it does not re-fetch. (If the agent omits it, the CLI fetches the same value itself — correct but one extra round-trip.)

create-limit

100010 ORDER_AMOUNT_TOO_SMALL

1. Fetch the from-token price (USD):
   • If from-token is a well-known stablecoin (USDT / USDC / USDG / USDe / DAI / FDUSD / ...): assume

     from_price ≈ 1.0

     without an HTTP call.
   • Otherwise: call

     onchainos market price --chain <chain> --address <from_token>

     , read

     data[0].price

     as

     from_price

     .
2. Compute USD value:

   usd_value = from_amount × from_price

   .
3. If

   usd_value < 1.0

   :
   • Compute

     min_from_amount = ceil(1.0 ÷ from_price)

     rounded up to a reasonable display precision for the token (e.g. whole units when

     from_price ≥ 0.1

     ; 2-4 significant digits otherwise).
   • Surface exactly this single canonical line to the user, with no extra prose — no USD-value math, no $1 threshold mention, no echo of the user's original amount, no follow-up sentence, no apology:

     Minimum order amount: <min_from_amount> <from_symbol>

     Translate the prefix at output time per §Display labels & output language (e.g. for a Chinese user the agent renders the same fact in Chinese). The structure stays single-line:

     <localised prefix> <min_from_amount> <from_symbol>

     .
   • STOP. Do NOT render Step 1. Do NOT call the CLI. Wait for the user to provide a larger

     --amount

     , then re-run Step 0 from the top.
4. If

   usd_value ≥ 1.0

   : carry

   from_price

   forward (Step 1's "Value" column reuses it; no need to re-fetch) and proceed to Step 1.

• from_price = 0.10

  ,

  usd_value = 1 × 0.10 = 0.10 < 1.0

  → fail

• min_from_amount = ceil(1.0 / 0.10) = 10

• Output:

  Minimum order amount: 10 OKB

• Stop. No Step 1, no extra prose.

--chain-id

strategyType

Buy Dip

Take Profit

Stop Loss

Buy Above

USDC

10

--amount

ARB

$0.10

Default 15%

User-specified X%

• Buy direction (BUY_DIP / CHASE_HIGH):

  • Estimated Amount

    =

    from_amount

    ÷

    trigger_price

    (in units of to-token)

  • Value

    =

    from_amount

    ×

    from_token_USD_price

    (if from is a stablecoin, ≈

    from_amount

    )
• Sell direction (TAKE_PROFIT / STOP_LOSS):

  • Estimated Amount

    =

    from_amount

    ×

    trigger_price

    (in units of to-token, usually a stablecoin)

  • Value

    =

    from_amount

    ×

    trigger_price

    (if to is a stablecoin, equals Estimated Amount)

• User did NOT mention slippage in the conversation → display

  Slippage: Default 15%

  , and omit

  --slippage

  on the CLI call (the CLI default is 15).
• User explicitly said "slippage X%" / "use X% slippage" / similar → display

  Slippage: User-specified X%

  , and pass

  --slippage X

  on the CLI call.

strategyType

status

1. Chain: Arbitrum
2. Order Type: Buy Dip
3. From: USDC 10
4. To:
   - Symbol: ARB
   - Trigger Price: $0.10
   - Estimated Amount: 100 ARB
   - Value: $10
5. Slippage: 15% (default)

If the trigger condition is not met within 7 days, this order auto-expires.

Reply confirm / change / cancel.

If the trigger condition is not met within 7 days, this order auto-expires.

• User says "confirm" / "yes" / "submit" → call

  onchainos strategy create-limit ...

  with the

  --type

  derived in Step 1.
• User says "change amount = 5" / "set trigger to 0.08" / similar → update the corresponding field and re-render Step 1 for another confirmation.
• User says "cancel" / "abort" → do NOT call the CLI; acknowledge that the order was discarded.

Hard constraints:

1. Never call

   strategy create-limit

   until the user has explicitly confirmed.

2. Estimated Amount

   /

   Value

   are agent-side estimates derived from

   trigger_price

   , not BE quotes. The realised fill amount is decided at BE execution time by slippage and aggregator routing; the agent must not present these estimates as "actual fill amounts".

3. --trigger-price

   is a USD price. The agent must make this clear to the user to avoid confusion with "exchange rate = X from-token per 1 to-token".
4. Never render Step 1 when Step 0's USD-value check fails. Output the single-line minimum-amount warning instead and stop — the user must restart with a larger

   --amount

   .

onchainos strategy list \
  [--order-id <id>] \
  [--status active,suspended,...] \
  [--chain-id 1,501] \
  [--token <address>] \
  [--limit <int>] \
  [--cursor <string>] \
  [--strategy-mode 7]

• Single order: pass

  --order-id <id>

  → GET

  openOrderDetail

  (returns full order shape).
• Page query: omit

  --order-id

  → POST

  getOpenOrder

  . The active wallet's addresses are auto-supplied; pass

  --limit

  (max 100, default 100) and

  --cursor

  from the previous response's

  nextCursor

  for pagination.

--status

--chain-id

orderStatusList

chainIdList

--token

tokenAddressList: List<String>

tokenAddress: String

list

data.list[].status

--status

expired

cancelling

cancelled

failed

processing

trading

completed

creating

active

suspended

SPEEDING_UP (-4) was removed 2026-05-08 — not surfaced and not a valid filter option. The Display label column is the only user-facing string; see §Display labels & output language. CLI's

statusLabel

already returns the display label, so just translate it per conversation language.

{-3, 0, 2, 3, 4}

{-7, -2, -1, 1}

--status

orderStatusList=[-3, 0, 2, 3, 4]

--status

--status

completed

cancelled

failed

expired

active,suspended,creating,trading,cancelling,completed,cancelled,failed,expired

--status

--status

4

active

suspended

processing

creating

cancelling

cancelled

completed

failed

expired

1. Run

   onchainos strategy list --limit 10

   (no

   --status

   ) — the CLI puts

   orderStatusList=[-3, 0, 2, 3, 4]

   (non-terminal set) into the request body; BE applies the filter server-side and returns only matching orders. Always pass

   --limit 10

   for general "show my orders" queries; full pagination is opt-in via "next page" follow-up.
2. Render the response

   data.list

   as a Markdown table with exactly these 8 columns (locked):

   Order id	Order Status	Order Type	Estimated Amount	To Token addr	Value	Trigger price	Expire after

   Per-row mapping (order matters, no extra columns):

   Column	Source	Notes
   Order id	data.list[i].orderId	—
   Order Status	data.list[i].statusLabel (Display label per the § status table)	Translate per §Display labels & output language
   Order Type	Display label per the § strategyType table, derived from data.list[i].strategyType integer ( 2 → Buy Dip / 3 → Take Profit / 4 → Stop Loss / 5 → Buy Above )	Translate per §Display labels & output language
   Estimated Amount	data.list[i].toToken.tokenAmount + + data.list[i].toToken.tokenSymbol	e.g. 0.2 SOL
   To Token addr	data.list[i].toToken.tokenContractAddress , truncated to first-6 + last-4	EVM: 0x1234...cdef ( 0x + 4 chars + ... + 4 chars); Solana base58: first 6 + ... + last 4
   Value	data.list[i].toToken.tokenUsd , formatted as <n> USD	e.g. 16 USD (round or 2-decimal, follow BE precision)
   Trigger price	data.list[i].triggerInfo.triggerPrice , prefixed with $	e.g. $80 ; if empty (trigger-rate path, not used in Phase 1) display data.list[i].triggerInfo.triggerRate
   Expire after	data.list[i].expireTime (13-digit ms UTC), converted to the user's current timezone, formatted MM/DD/YYYY HH:MM:SS	e.g. 05/15/2026 17:50:49 (semantically = createTime + 7 days by default)

   Sample row (Solana SOL→USDC take_profit):

   | 17262791359882688 | Active | Take Profit | 0.2 SOL | 9xQeWv...vEjz | 16 USD | $80 | 05/15/2026 17:50:49 |

   Address shortening rules:
   • EVM (

     0x

     prefix): first 6 chars (

     0x

     + 4) +

     ...

     + last 4 chars
   • Solana / other non-prefixed base58: first 6 +

     ...

     + last 4
   • Strings shorter than 10 chars: do not truncate; display verbatim
   Expire timezone conversion:

   • expireTime

     from BE is a UTC millisecond timestamp
   • The agent must convert to the user's current local timezone at render time (e.g. JS

     Intl.DateTimeFormat

     , Rust

     chrono::Local

     , or equivalent)
   • Fixed format:

     MM/DD/YYYY HH:MM:SS

     , 24-hour clock
3. After the table, append a single combined reminder covering pagination + status filter. Include the pagination line only when

   nextCursor

   is non-empty.
   Canonical EN; the agent translates per §Display labels & output language:

   Showing live orders by default (10 per page).

   • Reply "next page" to load more.
   • To filter by a specific state, ask for orders by their Display label — e.g.

     Completed

     ,

     Cancelled

     ,

     Failed

     ,

     Expired

     . Example: "show my completed orders" → I'll re-query with that filter.

   If

   nextCursor

   is empty (no more pages), drop the "next page" bullet and keep only the status-filter bullet.
4. If the user replies "next page" / similar, re-run with

   --limit 10 --cursor <nextCursor>

   carrying the previous response's

   nextCursor

   . Render with the same table format.
5. If the user names a specific status (any of the 9 Display labels), re-run as

   list --limit 10 --status <label>

   (single value) and render the same table. Drop the status-filter bullet from the reminder; keep the pagination bullet if

   nextCursor

   is non-empty.

minAmount = <N>

0.001

onchainos swap execute

onchainos wallet login

list

cancel --all

cancel --order-id

--amount

onchainos wallet balance

Match by integer code, not msg string. The msg text is for humans and may change.

• Latest entry wins. Render the most recent

  executionHistoryList[]

  item; older entries are historical context only.
• Repeated same code. When the same code recurs every ~10s without a

  txHash

  appearing, the engine is in a soft retry loop. Surface the latest

  message

  , mention the repeat count ("this is the 5th

  No quote due to low liquidity

  event"), and ask the user whether to wait, cancel, or adjust parameters.

• terminal=true

  codes. Stop polling immediately and surface to the user; engine will not retry.

• terminal=false

  codes that repeat 3+ times. Treat as user-actionable: suggest changing slippage / amount / pair / chain, or cancelling.
• Code

  0

  with

  txHash

  . Trade succeeded; surface

  txHash

  and explorer link.

create-limit

cancel

resume

estimatedWaitTime

1. Run the subcommand.
2. Sleep 3 seconds. A single fixed 3-second wait is enough across all supported chains in practice — empirically validated 2026-05-11. Do not read

   estimatedWaitTime

   from the response and sleep by that value; the BE-supplied number is conservative and inflates the wait unnecessarily (especially on ETH-class chains).
3. Re-query with

   onchainos strategy list --order-id <orderId>

   to read the final state.
4. If still pending after the first re-query, surface the partial state to the user; do not loop indefinitely.

Do not run a long polling loop. The CLI is not designed for that; the BE backs the order finalisation off-chain and one targeted re-query is sufficient in normal conditions.

• On

  create-limit

  or

  resume

  , if the BE returns code

  60018

  (

  UPGRADE_REQUIRED

  ), the CLI calls the Agentic Wallet attestation endpoints (

  getAttestDocHex

  → ed25519 sign with the session key →

  registerTeeInfo

  ), then retries the original op once. On success the user sees

  Trader Mode activated.

  followed by the normal command output.
• The agent should not ask the user to "activate Trader Mode" first — the CLI does it on demand.
• If activation fails, the original command aborts and the activation error is surfaced. Suggest

  onchainos wallet status

  to inspect the session.