MCP 服务器
Wicked MCP
io.github.choaticpixels/wicked-mcp
此 MCP 可以做什么
Provides wallet-based agent identity, reputation, signed persistent memory, tool reliability registry, claim verification, and crypto market data.
功能
工具
30
工具
identity_get_challenge
Request a real, time-boxed agent-liveness challenge from Wicked
Identity for an already-registered wallet. Read the returned
`instructions` carefully â it's a constrained-generation task (exact
word count + a letter-sum constraint) with a tight time budget
(`time_budget_seconds`, default 12s); submit your answer via
identity_submit_response before `expires_at`.
Requires a real signature over identity_get_nonce's message â this tool
does not sign anything itself.
Args:
wallet: Your agent's 0x wallet address on Base â must already be
registered via identity_register.
nonce: The nonce from a fresh identity_get_nonce call.
signature: The 0x-prefixed ECDSA signature over that nonce's message.
输入模式
{'type': 'object', 'title': 'identity_get_challengeArguments', 'required': ['wallet', 'nonce', 'signature'], 'properties': {'nonce': {'type': 'string', 'title': 'Nonce'}, 'wallet': {'type': 'string', 'title': 'Wallet'}, 'signature': {'type': 'string', 'title': 'Signature'}}}
输出模式
{'type': 'object', 'title': 'identity_get_challengeDictOutput', 'additionalProperties': True}
identity_get_nonce
Get a one-time nonce + message to sign for Wicked Identity â the
first step before identity_register, identity_get_challenge, or
identity_submit_response, each of which needs a FRESH nonce (single-use,
short TTL â don't reuse one across calls).
Args:
wallet: Your agent's 0x wallet address on Base.
输入模式
{'type': 'object', 'title': 'identity_get_nonceArguments', 'required': ['wallet'], 'properties': {'wallet': {'type': 'string', 'title': 'Wallet'}}}
输出模式
{'type': 'object', 'title': 'identity_get_nonceDictOutput', 'additionalProperties': True}
identity_jwks
Get Wicked Identity's public JWKS (RS256 keys) â use this to verify
an assertion_token's signature locally with a standard JWT library,
independent of calling identity_status. Public, free, no key needed.
输入模式
{'type': 'object', 'title': 'identity_jwksArguments', 'properties': {}}
输出模式
{'type': 'object', 'title': 'identity_jwksDictOutput', 'additionalProperties': True}
identity_register
Register your wallet with Wicked Identity â a lightweight identity
record, no stake required. Requires a real signature over the exact
message identity_get_nonce returned; this tool does not sign anything
itself.
Args:
wallet: Your agent's 0x wallet address on Base.
nonce: The nonce from a fresh identity_get_nonce call.
signature: The 0x-prefixed ECDSA signature over that nonce's message.
输入模式
{'type': 'object', 'title': 'identity_registerArguments', 'required': ['wallet', 'nonce', 'signature'], 'properties': {'nonce': {'type': 'string', 'title': 'Nonce'}, 'wallet': {'type': 'string', 'title': 'Wallet'}, 'signature': {'type': 'string', 'title': 'Signature'}}}
输出模式
{'type': 'object', 'title': 'identity_registerDictOutput', 'additionalProperties': True}
identity_status
Check whether a wallet currently holds a valid, unexpired Wicked
Identity assertion â the read-only check any third party can run on
someone else's wallet, no signature needed.
Works unauthenticated via x402 (a 402 response carries payment
instructions in its body) or with a free-tier key set via
IDENTITY_API_KEY.
Args:
wallet: The 0x wallet address to check.
输入模式
{'type': 'object', 'title': 'identity_statusArguments', 'required': ['wallet'], 'properties': {'wallet': {'type': 'string', 'title': 'Wallet'}}}
输出模式
{'type': 'object', 'title': 'identity_statusDictOutput', 'additionalProperties': True}
identity_submit_response
Submit your answer to a Wicked Identity challenge. Scored honestly as
pass / fail / inconclusive â never forced to a binary result. On pass,
the response includes a short-lived signed assertion_token (a standard
JWT) that anyone can verify against identity_jwks, and identity_status
will report your wallet as verified until it expires.
Requires a real signature over identity_get_nonce's message â this tool
does not sign anything itself. Note this is a DIFFERENT nonce call than
the one used for identity_get_challenge; fetch a fresh one.
Args:
wallet: Your agent's 0x wallet address on Base.
nonce: The nonce from a fresh identity_get_nonce call.
signature: The 0x-prefixed ECDSA signature over that nonce's message.
challenge_id: The challenge_id from identity_get_challenge.
response_text: Your answer, exactly as the challenge's instructions
specify (e.g. raw lowercase words, no extra commentary).
输入模式
{'type': 'object', 'title': 'identity_submit_responseArguments', 'required': ['wallet', 'nonce', 'signature', 'challenge_id', 'response_text'], 'properties': {'nonce': {'type': 'string', 'title': 'Nonce'}, 'wallet': {'type': 'string', 'title': 'Wallet'}, 'signature': {'type': 'string', 'title': 'Signature'}, 'challenge_id': {'type': 'string', 'title': 'Challenge Id'}, 'response_text': {'type': 'string', 'title': 'Response Text'}}}
输出模式
{'type': 'object', 'title': 'identity_submit_responseDictOutput', 'additionalProperties': True}
memory_delete
PERMANENTLY delete a memory (hard delete; right-to-be-forgotten). The
content is not recoverable; only an audit row (no content) is kept. The
default scope "chain" removes EVERY version of the memory; "version"
removes only this one. Call memory_prepare with operation "delete" and
the same arguments first.
Args:
wallet: Your agent's 0x wallet address.
memory_id: UUID of any version in the chain.
timestamp: From memory_prepare.
nonce: From memory_prepare (single use).
signature: Your wallet's signature over memory_prepare's message_to_sign.
scope: "chain" (all versions, default) or "version" (just this row).
reason: Optional note recorded in the audit log (max 200 chars).
输入模式
{'type': 'object', 'title': 'memory_deleteArguments', 'required': ['wallet', 'memory_id', 'timestamp', 'nonce', 'signature'], 'properties': {'nonce': {'type': 'string', 'title': 'Nonce'}, 'scope': {'type': 'string', 'title': 'Scope', 'default': 'chain'}, 'reason': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Reason', 'default': None}, 'wallet': {'type': 'string', 'title': 'Wallet'}, 'memory_id': {'type': 'string', 'title': 'Memory Id'}, 'signature': {'type': 'string', 'title': 'Signature'}, 'timestamp': {'type': 'string', 'title': 'Timestamp'}}}
输出模式
{'type': 'object', 'title': 'memory_deleteDictOutput', 'additionalProperties': True}
memory_deletions
Your deletion audit log (ids, times, reasons; never content). Call
memory_prepare with operation "deletions" first.
Args:
wallet: Your agent's 0x wallet address.
timestamp: From memory_prepare.
nonce: From memory_prepare (single use).
signature: Your wallet's signature over memory_prepare's message_to_sign.
输入模式
{'type': 'object', 'title': 'memory_deletionsArguments', 'required': ['wallet', 'timestamp', 'nonce', 'signature'], 'properties': {'nonce': {'type': 'string', 'title': 'Nonce'}, 'wallet': {'type': 'string', 'title': 'Wallet'}, 'signature': {'type': 'string', 'title': 'Signature'}, 'timestamp': {'type': 'string', 'title': 'Timestamp'}}}
输出模式
{'type': 'object', 'title': 'memory_deletionsDictOutput', 'additionalProperties': True}
memory_get
Fetch one of your memories by id. Another wallet's id (or a deleted
one) returns the same 404 as an id that never existed. Call
memory_prepare with operation "get" and params {"memory_id": ...} first.
Args:
wallet: Your agent's 0x wallet address.
memory_id: The memory's UUID.
timestamp: From memory_prepare.
nonce: From memory_prepare (single use).
signature: Your wallet's signature over memory_prepare's message_to_sign.
输入模式
{'type': 'object', 'title': 'memory_getArguments', 'required': ['wallet', 'memory_id', 'timestamp', 'nonce', 'signature'], 'properties': {'nonce': {'type': 'string', 'title': 'Nonce'}, 'wallet': {'type': 'string', 'title': 'Wallet'}, 'memory_id': {'type': 'string', 'title': 'Memory Id'}, 'signature': {'type': 'string', 'title': 'Signature'}, 'timestamp': {'type': 'string', 'title': 'Timestamp'}}}
输出模式
{'type': 'object', 'title': 'memory_getDictOutput', 'additionalProperties': True}
memory_history
Full version chain for a memory, oldest first, with each version's
`valid_from` / `valid_until` / `superseded_by`. Works from any version's
id. Call memory_prepare with operation "history" first.
Args:
wallet: Your agent's 0x wallet address.
memory_id: UUID of any version in the chain.
timestamp: From memory_prepare.
nonce: From memory_prepare (single use).
signature: Your wallet's signature over memory_prepare's message_to_sign.
输入模式
{'type': 'object', 'title': 'memory_historyArguments', 'required': ['wallet', 'memory_id', 'timestamp', 'nonce', 'signature'], 'properties': {'nonce': {'type': 'string', 'title': 'Nonce'}, 'wallet': {'type': 'string', 'title': 'Wallet'}, 'memory_id': {'type': 'string', 'title': 'Memory Id'}, 'signature': {'type': 'string', 'title': 'Signature'}, 'timestamp': {'type': 'string', 'title': 'Timestamp'}}}
输出模式
{'type': 'object', 'title': 'memory_historyDictOutput', 'additionalProperties': True}
memory_prepare
Step 1 of every Wicked Memory call: get the exact message to sign.
Wicked Memory is persistent, wallet-scoped memory for agents (store,
semantic search, versioned history, hard delete). Each call needs a fresh
wallet signature; this server never signs or holds a key.
Sign `message_to_sign` with your wallet (EIP-191 personal_sign), then
call the matching tool (memory_store, memory_search, ...) with the SAME
arguments plus the returned `timestamp`, `nonce` and your `signature`.
The timestamp must be within 5 minutes and the nonce is single-use, so
prepare again for every call.
Args:
operation: One of "store", "search", "get", "update", "history",
"delete", "deletions".
wallet: Your agent's 0x wallet address (your memories are scoped to it).
params: The same arguments you will pass to the tool, by name, e.g.
{"content": "...", "tags": ["a"]} for store, {"q": "..."} for
search, {"memory_id": "<uuid>"} for get/update/history/delete.
输入模式
{'type': 'object', 'title': 'memory_prepareArguments', 'required': ['operation', 'wallet'], 'properties': {'params': {'anyOf': [{'type': 'object', 'additionalProperties': True}, {'type': 'null'}], 'title': 'Params', 'default': None}, 'wallet': {'type': 'string', 'title': 'Wallet'}, 'operation': {'type': 'string', 'title': 'Operation'}}}
输出模式
{'type': 'object', 'title': 'memory_prepareDictOutput', 'additionalProperties': True}
memory_search
Semantic search over YOUR memories (scoped to `wallet`; no other
wallet's memories are ever returned). Ranked by real cosine similarity
computed at query time (`score`). Call memory_prepare with operation
"search" and these same arguments first.
Access: this is the one metered endpoint. With a free-tier key
(MEMORY_API_KEY, if this server has one) it is free and rate-limited;
otherwise it returns http_status 402 with x402 payment instructions under
`payment_required` (0.001 USDC on Base). To pay: build and sign the x402
payment from `payment_required`, then call this tool AGAIN with the SAME
arguments (same timestamp/nonce/signature, which a 402 does not consume)
plus `payment_signature`.
By default only currently-active versions are searched; use `as_of` to
search memory as it stood at a past instant, or include_superseded=true
for every version.
Args:
wallet: Your agent's 0x wallet address.
q: What to look for, in natural language.
timestamp: From memory_prepare.
nonce: From memory_prepare (single use).
signature: Your wallet's signature over memory_prepare's message_to_sign.
limit: 1-50 results (default 10).
tags: Only memories having ALL of these tags.
created_after: ISO-8601 lower bound on creation time.
created_before: ISO-8601 upper bound on creation time.
as_of: ISO-8601 instant; search the versions valid at that moment.
include_superseded: Also search old, superseded versions.
payment_signature: x402 payment payload (base64) for the keyless path.
输入模式
{'type': 'object', 'title': 'memory_searchArguments', 'required': ['wallet', 'q', 'timestamp', 'nonce', 'signature'], 'properties': {'q': {'type': 'string', 'title': 'Q'}, 'tags': {'anyOf': [{'type': 'array', 'items': {'type': 'string'}}, {'type': 'null'}], 'title': 'Tags', 'default': None}, 'as_of': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'As Of', 'default': None}, 'limit': {'anyOf': [{'type': 'integer'}, {'type': 'null'}], 'title': 'Limit', 'default': None}, 'nonce': {'type': 'string', 'title': 'Nonce'}, 'wallet': {'type': 'string', 'title': 'Wallet'}, 'signature': {'type': 'string', 'title': 'Signature'}, 'timestamp': {'type': 'string', 'title': 'Timestamp'}, 'created_after': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Created After', 'default': None}, 'created_before': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Created Before', 'default': None}, 'payment_signature': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Payment Signature', 'default': None}, 'include_superseded': {'anyOf': [{'type': 'boolean'}, {'type': 'null'}], 'title': 'Include Superseded', 'default': None}}}
输出模式
{'type': 'object', 'title': 'memory_searchDictOutput', 'additionalProperties': True}
memory_store
Store a memory in Wicked Memory (free; fair-use rate limited). Call
memory_prepare with operation "store" and these same arguments first, sign
its message, then pass timestamp/nonce/signature here. A real embedding is
computed at write time. Returns the stored memory (`data.id`, validity
window, ...).
Args:
wallet: Your agent's 0x wallet address (scopes the memory).
content: The text to remember (max 16 KB).
timestamp: From memory_prepare.
nonce: From memory_prepare (single use).
signature: Your wallet's signature over memory_prepare's message_to_sign.
tags: Up to 20 short tags for filtering.
metadata: Free-form JSON object (max 8 KB).
source: Which agent/session wrote this.
输入模式
{'type': 'object', 'title': 'memory_storeArguments', 'required': ['wallet', 'content', 'timestamp', 'nonce', 'signature'], 'properties': {'tags': {'anyOf': [{'type': 'array', 'items': {'type': 'string'}}, {'type': 'null'}], 'title': 'Tags', 'default': None}, 'nonce': {'type': 'string', 'title': 'Nonce'}, 'source': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Source', 'default': None}, 'wallet': {'type': 'string', 'title': 'Wallet'}, 'content': {'type': 'string', 'title': 'Content'}, 'metadata': {'anyOf': [{'type': 'object', 'additionalProperties': True}, {'type': 'null'}], 'title': 'Metadata', 'default': None}, 'signature': {'type': 'string', 'title': 'Signature'}, 'timestamp': {'type': 'string', 'title': 'Timestamp'}}}
输出模式
{'type': 'object', 'title': 'memory_storeDictOutput', 'additionalProperties': True}
memory_update
Update a memory WITHOUT overwriting it: creates a new version and
closes the old one (`valid_until` / `superseded_by`), so history is kept.
Omitted fields carry over. Returns `previous` and `current`. Only the
current version can be updated (409 otherwise, naming the current id).
Call memory_prepare with operation "update" and the same arguments first.
Args:
wallet: Your agent's 0x wallet address.
memory_id: UUID of the CURRENT version to supersede.
timestamp: From memory_prepare.
nonce: From memory_prepare (single use).
signature: Your wallet's signature over memory_prepare's message_to_sign.
content: New text (re-embedded). Provide at least one of content/tags/metadata/source.
tags: Replacement tags.
metadata: Replacement metadata object.
source: New source label.
输入模式
{'type': 'object', 'title': 'memory_updateArguments', 'required': ['wallet', 'memory_id', 'timestamp', 'nonce', 'signature'], 'properties': {'tags': {'anyOf': [{'type': 'array', 'items': {'type': 'string'}}, {'type': 'null'}], 'title': 'Tags', 'default': None}, 'nonce': {'type': 'string', 'title': 'Nonce'}, 'source': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Source', 'default': None}, 'wallet': {'type': 'string', 'title': 'Wallet'}, 'content': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Content', 'default': None}, 'metadata': {'anyOf': [{'type': 'object', 'additionalProperties': True}, {'type': 'null'}], 'title': 'Metadata', 'default': None}, 'memory_id': {'type': 'string', 'title': 'Memory Id'}, 'signature': {'type': 'string', 'title': 'Signature'}, 'timestamp': {'type': 'string', 'title': 'Timestamp'}}}
输出模式
{'type': 'object', 'title': 'memory_updateDictOutput', 'additionalProperties': True}
registry_featured_tools
Get Wicked Registry's top scored active tools. Public, free, no key
or payment required â a quick look before calling registry_search_tools
with a specific filter.
输入模式
{'type': 'object', 'title': 'registry_featured_toolsArguments', 'properties': {}}
输出模式
{'type': 'object', 'title': 'registry_featured_toolsDictOutput', 'additionalProperties': True}
registry_register_tool
Register a tool you own with Wicked Registry so it starts getting
real synthetic checks and a live reliability score.
Requires a real signature over the exact message:
"Wicked Registry â Tool Registration\nname: {name}\nendpoint: {endpoint_url}\ntimestamp: {timestamp}"
signed by owner_wallet. This tool does not sign anything itself â you
(or your agent's own wallet) must produce that signature first; a
client-asserted wallet with no valid signature is rejected.
Args:
name: Tool name.
endpoint_url: The tool's live, publicly reachable endpoint â
checked on an interval once registered.
protocol: One of "x402", "mcp", "rest".
category: Free-text category, e.g. "data", "crypto".
description: What the tool does.
owner_wallet: The 0x wallet address that signed the registration message.
signature: The 0x-prefixed ECDSA signature over the registration message.
timestamp: ISO 8601 timestamp used in the signed message â must be
within 10 minutes of the call.
schema_url: Optional URL to a JSON Schema the tool's response
validates against â enables the schema-conformance score
component. Omit if the tool doesn't publish one.
输入模式
{'type': 'object', 'title': 'registry_register_toolArguments', 'required': ['name', 'endpoint_url', 'protocol', 'category', 'description', 'owner_wallet', 'signature', 'timestamp'], 'properties': {'name': {'type': 'string', 'title': 'Name'}, 'category': {'type': 'string', 'title': 'Category'}, 'protocol': {'type': 'string', 'title': 'Protocol'}, 'signature': {'type': 'string', 'title': 'Signature'}, 'timestamp': {'type': 'string', 'title': 'Timestamp'}, 'schema_url': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Schema Url', 'default': None}, 'description': {'type': 'string', 'title': 'Description'}, 'endpoint_url': {'type': 'string', 'title': 'Endpoint Url'}, 'owner_wallet': {'type': 'string', 'title': 'Owner Wallet'}}}
输出模式
{'type': 'object', 'title': 'registry_register_toolDictOutput', 'additionalProperties': True}
registry_report_tool
File a real complaint about a tool's reliability or behavior with
Wicked Registry. Public, rate-limited, no auth required. Reviewed
manually by an admin â filing a report does not auto-suspend the tool.
Args:
tool_id: The tool's Wicked Registry id.
reporter: Your wallet address, or another best-effort identifier.
reason: One of "downtime", "incorrect_response", "scam_or_abuse",
"schema_violation", "other".
evidence: Optional free-text note or URL supporting the report.
输入模式
{'type': 'object', 'title': 'registry_report_toolArguments', 'required': ['tool_id', 'reporter', 'reason'], 'properties': {'reason': {'type': 'string', 'title': 'Reason'}, 'tool_id': {'type': 'string', 'title': 'Tool Id'}, 'evidence': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Evidence', 'default': None}, 'reporter': {'type': 'string', 'title': 'Reporter'}}}
输出模式
{'type': 'object', 'title': 'registry_report_toolDictOutput', 'additionalProperties': True}
registry_search_tools
Search Wicked Registry for x402/MCP tools by real, live reliability
data (uptime, latency, schema conformance â never fabricated).
Works unauthenticated via x402 (a 402 response carries payment
instructions in its body) or with a free-tier key set via
REGISTRY_API_KEY.
Args:
category: Filter by category, e.g. "data", "crypto".
protocol: Filter by protocol â "x402", "mcp", or "rest".
min_score: Minimum composite score (0-100). Tools with no check
history yet are excluded once this is set, rather than treated
as a 0.
sort: One of "score" (default), "latency", "recency".
limit: Max results, 1-200. Defaults to 50.
输入模式
{'type': 'object', 'title': 'registry_search_toolsArguments', 'properties': {'sort': {'type': 'string', 'title': 'Sort', 'default': 'score'}, 'limit': {'type': 'integer', 'title': 'Limit', 'default': 50}, 'category': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Category', 'default': None}, 'protocol': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Protocol', 'default': None}, 'min_score': {'anyOf': [{'type': 'number'}, {'type': 'null'}], 'title': 'Min Score', 'default': None}}}
输出模式
{'type': 'object', 'title': 'registry_search_toolsDictOutput', 'additionalProperties': True}
registry_tool_badge
Get a tool's current composite reliability score. Public, free, no
key or payment required â the same badge a tool owner embeds on their
own docs.
Args:
tool_id: The tool's Wicked Registry id.
输入模式
{'type': 'object', 'title': 'registry_tool_badgeArguments', 'required': ['tool_id'], 'properties': {'tool_id': {'type': 'string', 'title': 'Tool Id'}}}
输出模式
{'type': 'object', 'title': 'registry_tool_badgeDictOutput', 'additionalProperties': True}
registry_tool_detail
Get one tool's full score breakdown, component history, and recent
real health checks from Wicked Registry.
Works unauthenticated via x402, or with a free-tier key set via
REGISTRY_API_KEY.
Args:
tool_id: The tool's Wicked Registry id, from registry_search_tools
or registry_featured_tools.
输入模式
{'type': 'object', 'title': 'registry_tool_detailArguments', 'required': ['tool_id'], 'properties': {'tool_id': {'type': 'string', 'title': 'Tool Id'}}}
输出模式
{'type': 'object', 'title': 'registry_tool_detailDictOutput', 'additionalProperties': True}
reputation_query
Search/rank registered agents by tier, stake, reputation, or slash
history. A safe, parameterized alternative to raw SQL â every filter is
a typed query param, there's no query-injection surface.
Args:
tier: Exact tier name to filter to, e.g. "gold".
min_stake / max_stake: Active-stake bounds in USDC.
min_reputation: Minimum reputation score.
founding_member: True to only show founding members.
min_slash_count: Minimum number of slash events (e.g. 1 to find agents that have ever been slashed).
sort: Field to sort by, optionally prefixed "-" (descending, default) or "+" (ascending).
One of stake_amount, reputation_score, slash_count, registered_at.
limit: Max results, 1-200. Defaults to 20.
输入模式
{'type': 'object', 'title': 'reputation_queryArguments', 'properties': {'sort': {'type': 'string', 'title': 'Sort', 'default': '-stake_amount'}, 'tier': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Tier', 'default': None}, 'limit': {'type': 'integer', 'title': 'Limit', 'default': 20}, 'max_stake': {'anyOf': [{'type': 'number'}, {'type': 'null'}], 'title': 'Max Stake', 'default': None}, 'min_stake': {'anyOf': [{'type': 'number'}, {'type': 'null'}], 'title': 'Min Stake', 'default': None}, 'min_reputation': {'anyOf': [{'type': 'number'}, {'type': 'null'}], 'title': 'Min Reputation', 'default': None}, 'founding_member': {'anyOf': [{'type': 'boolean'}, {'type': 'null'}], 'title': 'Founding Member', 'default': None}, 'min_slash_count': {'anyOf': [{'type': 'integer'}, {'type': 'null'}], 'title': 'Min Slash Count', 'default': None}}}
输出模式
{'type': 'object', 'title': 'reputation_queryDictOutput', 'additionalProperties': True}
reputation_statement
Get an agent's full position + real event ledger (stakes, unstakes,
withdrawals, slashes) from Wicked Reputation's own records. Does NOT
include WickedAPI call volume or fees paid â see the `note` field in the
response for exactly what this does and doesn't cover.
Args:
wallet: The agent's 0x wallet address on Base.
输入模式
{'type': 'object', 'title': 'reputation_statementArguments', 'required': ['wallet'], 'properties': {'wallet': {'type': 'string', 'title': 'Wallet'}}}
输出模式
{'type': 'object', 'title': 'reputation_statementDictOutput', 'additionalProperties': True}
reputation_status
Get an agent's current reputation: tier, active/unbonding stake,
reputation score, slash count, founding-member badge. Returns
http_status 404 for a wallet that has never registered â treat that the
same as tier "unranked", not an error.
Args:
wallet: The agent's 0x wallet address on Base.
输入模式
{'type': 'object', 'title': 'reputation_statusArguments', 'required': ['wallet'], 'properties': {'wallet': {'type': 'string', 'title': 'Wallet'}}}
输出模式
{'type': 'object', 'title': 'reputation_statusDictOutput', 'additionalProperties': True}
reputation_tiers
Get Wicked Reputation's live tier thresholds and benefits
(min stake required, rate-limit multiplier, fee-discount %). Read at
call time rather than assumed â thresholds are admin-configurable.
输入模式
{'type': 'object', 'title': 'reputation_tiersArguments', 'properties': {}}
输出模式
{'type': 'object', 'title': 'reputation_tiersDictOutput', 'additionalProperties': True}
reputation_transparency
Get Wicked Reputation's live transparency numbers: treasury address,
custody model, block-explorer link, unbonding cooldown, total active
stake, total agents, total slash events, founding-slot counts.
输入模式
{'type': 'object', 'title': 'reputation_transparencyArguments', 'properties': {}}
输出模式
{'type': 'object', 'title': 'reputation_transparencyDictOutput', 'additionalProperties': True}
sanity_check
Verify a claim before you act on it or hand it downstream. Every
verdict comes from a real model inference run at request time (HHEM
entailment) â never cached or guessed.
Two modes:
- "grounded" (default): is the claim faithful to the `context` you
supply? `context` is REQUIRED. This is a scoreable question.
- "open": is the claim consistent with live web evidence found right
now (real Tavily search, then the same model)? `context` is ignored.
This is consistency with current web content, NOT objective truth â
a claim can be well-supported by the web and still be wrong.
Reading the result (in `data`):
- `verdict`: "supported" (consistent with the source), "contradicted"
(clearly inconsistent), "unsupported" (not supported but not clearly
contradicted â treat as unverified), or "insufficient_evidence"
(open mode found nothing usable; says nothing about whether the
claim is true).
- `confidence_score`: the raw consistency score in [0, 1] â near 1 =
consistent with the source, near 0 = inconsistent. It is NOT
confidence-in-the-verdict: a "contradicted" verdict has a score near
0. It is 0.0 for "insufficient_evidence".
- `evidence`: the audit trail (context snippet, or web citations).
Suggested policy: act on "supported"; treat "unsupported" and
"insufficient_evidence" as unverified; reject or regenerate on
"contradicted".
Access: the public mcp.wickedapi.com server uses a shared, rate-limited
key, so this returns real verdicts directly -- but the shared budget is
small (about 20 requests/minute and 30 open-mode checks/day across ALL
users; grounded mode is not counted against the daily limit). An
http_status 429 means that shared budget is used up: retry after the
Retry-After seconds, run this server locally with your own
SANITY_API_KEY, or pay per call via x402 directly. A server with no key
configured returns an http_status 402 carrying x402 payment instructions
under `payment_required` instead.
Args:
claim: The single statement to verify (max 5,000 characters).
context: Source text the claim must be faithful to â a string or a
list of strings. Required for mode "grounded".
mode: "grounded" (check against `context`) or "open" (check against
live web evidence).
输入模式
{'type': 'object', 'title': 'sanity_checkArguments', 'required': ['claim'], 'properties': {'mode': {'type': 'string', 'title': 'Mode', 'default': 'grounded'}, 'claim': {'type': 'string', 'title': 'Claim'}, 'context': {'anyOf': [{'type': 'string'}, {'type': 'array', 'items': {'type': 'string'}}, {'type': 'null'}], 'title': 'Context', 'default': None}}}
输出模式
{'type': 'object', 'title': 'sanity_checkDictOutput', 'additionalProperties': True}
sanity_check_batch
Verify several claims in one call â use this for a multi-sentence
output instead of calling sanity_check once per sentence. Split the
output into individual claims, pass the source as `context`, and every
claim is checked against it. One payment / one rate-limit hit covers the
whole batch, and grounded batches are scored far faster than N separate
calls.
Returns `data` as a list with one result per claim, in the same order
as `claims`; each result is shaped exactly like sanity_check's (see
that tool for how to read `verdict`, `confidence_score` and `evidence`).
In "open" mode a separate live web search runs per claim, so large open
batches are slow.
Access: the public mcp.wickedapi.com server uses a shared, rate-limited
key, so this returns real verdicts directly -- but the shared budget is
small (about 20 requests/minute and 30 open-mode checks/day across ALL
users; grounded mode is not counted against the daily limit). An
http_status 429 means that shared budget is used up: retry after the
Retry-After seconds, run this server locally with your own
SANITY_API_KEY, or pay per call via x402 directly. A server with no key
configured returns an http_status 402 carrying x402 payment instructions
under `payment_required` instead.
Args:
claims: 1-50 statements to verify, each up to 5,000 characters.
context: Shared source text for every claim â a string or a list of
strings. Required for mode "grounded".
mode: "grounded" (check against `context`) or "open" (check against
live web evidence).
输入模式
{'type': 'object', 'title': 'sanity_check_batchArguments', 'required': ['claims'], 'properties': {'mode': {'type': 'string', 'title': 'Mode', 'default': 'grounded'}, 'claims': {'type': 'array', 'items': {'type': 'string'}, 'title': 'Claims'}, 'context': {'anyOf': [{'type': 'string'}, {'type': 'array', 'items': {'type': 'string'}}, {'type': 'null'}], 'title': 'Context', 'default': None}}}
输出模式
{'type': 'object', 'title': 'sanity_check_batchDictOutput', 'additionalProperties': True}
wickedapi_funding_oi
Get perpetual-futures funding rate and open interest for a crypto symbol.
Args:
symbol: Base crypto symbol, e.g. BTC, ETH, SOL â mapped to the
<SYMBOL>USDT perpetual on Binance/Bybit.
输入模式
{'type': 'object', 'title': 'wickedapi_funding_oiArguments', 'required': ['symbol'], 'properties': {'symbol': {'type': 'string', 'title': 'Symbol'}}}
输出模式
{'type': 'object', 'title': 'wickedapi_funding_oiDictOutput', 'additionalProperties': True}
wickedapi_momentum
Get WickedAPI's momentum score for a symbol.
Args:
symbol: One of BTC, ETH, SOL, SUI, NVDA, IBIT, COIN, QQQ, TQQQ, DIA.
timeframe: One of 15m, 1h, 4h, 1D, 1W. Defaults to 1h.
输入模式
{'type': 'object', 'title': 'wickedapi_momentumArguments', 'required': ['symbol'], 'properties': {'symbol': {'type': 'string', 'title': 'Symbol'}, 'timeframe': {'type': 'string', 'title': 'Timeframe', 'default': '1h'}}}
输出模式
{'type': 'object', 'title': 'wickedapi_momentumDictOutput', 'additionalProperties': True}
wickedapi_price
Get the current price for a symbol (crypto or stock) from WickedAPI.
Args:
symbol: e.g. BTC, ETH, SUI, NVDA, SPY.
asset_class: "stock" or "crypto" â required for symbols outside the
curated list, optional otherwise.
输入模式
{'type': 'object', 'title': 'wickedapi_priceArguments', 'required': ['symbol'], 'properties': {'symbol': {'type': 'string', 'title': 'Symbol'}, 'asset_class': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Asset Class', 'default': None}}}
输出模式
{'type': 'object', 'title': 'wickedapi_priceDictOutput', 'additionalProperties': True}
历史
近期工具变更
已添加
memory_deletions
2026年10月4日 02:40
已添加
memory_delete
2026年10月4日 02:40
已添加
memory_history
2026年10月4日 02:40
已添加
memory_update
2026年10月4日 02:40
已添加
memory_get
2026年10月4日 02:40
已添加
memory_search
2026年10月4日 02:40
已添加
memory_store
2026年10月4日 02:40
已添加
memory_prepare
2026年10月4日 02:40
已添加
sanity_check_batch
2026年10月4日 02:40
已添加
sanity_check
2026年10月4日 02:40
已添加
identity_jwks
2026年10月4日 02:40
已添加
identity_status
2026年10月4日 02:40
已添加
identity_submit_response
2026年10月4日 02:40
已添加
identity_get_challenge
2026年10月4日 02:40
已添加
identity_register
2026年10月4日 02:40
已添加
identity_get_nonce
2026年10月4日 02:40
已添加
registry_register_tool
2026年10月4日 02:40
已添加
registry_report_tool
2026年10月4日 02:40
已添加
registry_tool_badge
2026年10月4日 02:40
已添加
registry_featured_tools
2026年10月4日 02:40
已添加
registry_tool_detail
2026年10月4日 02:40
已添加
registry_search_tools
2026年10月4日 02:40
已添加
reputation_transparency
2026年10月4日 02:40
已添加
reputation_tiers
2026年10月4日 02:40
已添加
reputation_query
2026年10月4日 02:40
已添加
reputation_statement
2026年10月4日 02:40
已添加
reputation_status
2026年10月4日 02:40
已添加
wickedapi_funding_oi
2026年10月4日 02:40
已添加
wickedapi_momentum
2026年10月4日 02:40
已添加
wickedapi_price
2026年10月4日 02:40
发现
类似的 MCP 服务器
AgentBIT
app.agentbit/mcp
Routes pay-per-call tools over x402 on Base, covering discovery, data conversion, company research, counterparty screening, domai…
MolTrust MCP Server
io.github.MoltyCel/moltrust-mcp-server
Provides agent identity, trust scoring, verifiable credentials, reputation, Sybil analysis, and on-chain mandate verification.
fresh-feeds — trending/new/changed MCP servers + x402 change-data API
io.github.foomworks/fresh-feeds
Indexes and verifies MCP servers and x402 services, provides registry change data, and assesses on-chain agent identity and payme…
AgentIndex
io.github.cognivis/agentindex
Discovers x402 services, evaluates their trustworthiness, and returns data with payment and verification receipts.
Delx Commerce — Pay-per-Result APIs for Agents
io.github.davidmosiah/delx-mcp-a2a
Provides pay-per-result APIs and agent protocol tools for continuity, multi-agent coordination, document OCR, and crypto-related …
SAP MCP Server
ai.oobeprotocol.sap.mcp/sap-mcp
Provides an MCP gateway connecting SAP-related tools with Solana, DeFi, decentralized identity, and x402 payments.
TheProtocol — Sovereign AI Agent Platform
cloud.theprotocol/registry
Manages sovereign AI agents, identities, contracts, payments, governance, exchanges, and cross-frame token transfers.
AIMEAT
io.aimeat/aimeat
Provides a self-hosted agent operating system with agent work delegation, access controls, federation, hooks, SSO, security admin…