app.aiconduit/conduit

Est-ce que app.aiconduit/conduit est sûr ?

Au dernier audit MCP Hub, le serveur MCP « app.aiconduit/conduit » est classé « Non encore testé ». Référencé dans l'annuaire. Seul le pentest comportemental fait foi, et il n'a pas encore été exécuté sur ce serveur. Le verdict est réévalué en continu ; consultez la fiche pour l'état à jour.

Méthode : MCP Hub analyse la description de chaque outil (couche 1) et, pour le verdict fort, exécute le serveur dans une micro-VM isolée avec leurres et réseau bloqué (couche 2). On publie des noms et des compteurs, jamais le contenu brut renvoyé par un outil.

Les 25 outils

• agent_authenticate≈ 630 tokens

  Two-step re-auth. Call with agent_id only → ES256-sign nonce → call again with nonce+signature. Prefer keys from ~/.conduit ({handle}.credentials.json or legacy credentials.json; CONDUIT_CREDENTIALS_PATH pins one file). After success, merge name/role/description into that file; never overwrite private_key; keep session_token in memory. Do not agent_create if identity files already exist unless the human asked for a new agent.

  Paramètres, retour & exemple (3)

  Paramètres

  • nonce · string · optionnel
    Challenge nonce from the first agent_authenticate call (omit to request one)
  • agent_id · string · requis
    Agent id from the chosen ~/.conduit credentials file, e.g. agt_...
  • signature · string · optionnel
    Signature of the nonce with the agent private key (omit with nonce to get challenge)

  Exemple d'appel

  {"name": "agent_authenticate", "arguments": {"agent_id": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"hint": "texte", "next": {}, "error": "texte", "nonce": "texte", "avatar": "texte", "detail": "texte", "handle": "texte", "persist": "texte", "agent_id": "texte", "mandates": [], "terminal": "texte", "challenge": "texte"}}

• agent_create≈ 1234 tokens

  Register an agent (ES256 P-256 public JWK JSON string + optional payment rails + destination). BEFORE: list ~/.conduit identity files — if any exist, reuse agent_id (do NOT register again unless the human asked for a new agent). AFTER: write persist.path (version, agent_id, public_key, private_key as JWKs; also handle, friendly_name, role_description, human_description; chmod 0600; write ~/.conduit/active). Then agent_update with default_destination+postcode before supply_search. Optional friendly_name / human_description / role_description / avatar only if the human stated them — never invent; omit avatar and the server assigns a random invader+gradient. Omit handle to auto-generate (e.g. parcel-watcher-12).

  Paramètres, retour & exemple (10)

  Paramètres

  • avatar · ? · optionnel
    Identity mark. Omit unless the human chose it — server assigns a random invader+gradient. { kind:"glyph", glyph:"invader"|emoji, tint?:invader color, gradient } or { kind:"image", src }. Tint only for invader. Do not store in the credentials file.
  • handle · string · optionnel
    Optional display handle (lowercase kebab); omit to auto-generate e.g. parcel-watcher-12
  • public_key · string · requis
    ES256 P-256 public JWK as a JSON string (kty=EC, crv=P-256, x, y). Not PEM, hex, or base64url-wrapped JSON. Keep the matching private JWK only in ~/.conduit/{handle}.credentials.json.
  • preferences · ? · optionnel
    Optional agent preference bag
  • friendly_name · string · optionnel
    Human-facing label (≠ handle), e.g. EU Restock Bot
  • payment_rails · array · optionnel
    Optional rail keys to seed, e.g. ["x402"]
  • business_profile · ? · optionnel
    Optional business profile metadata object
  • role_description · string · optionnel
    Self-instruction for this agent only, e.g. manage EU supplies
  • human_description · string · optionnel
    Description shown to humans on the org dashboard
  • default_destination · object · optionnel
    Where orders ship. postcode + country is enough for supply_search and delivery estimates; add address, locality and region so checkout links arrive prefilled.

  Exemple d'appel

  {"name": "agent_create", "arguments": {"public_key": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"hint": "texte", "next": {}, "error": "texte", "nonce": "texte", "avatar": "texte", "detail": "texte", "handle": "texte", "persist": "texte", "agent_id": "texte", "mandates": [], "terminal": "texte", "challenge": "texte"}}

• agent_notify≈ 1044 tokens

  Email Hub users in this agent’s linked organization (never addresses outside the org). to=all or { user_id } / { role } / { email } matching agent_organization members. Optional actions (Hub URLs only), severity, idempotency_key. Channel is email.

  Paramètres, retour & exemple (12)

  Paramètres

  • to · ? · requis
    Who to notify: "all", { user_id }, { user_ids }, { role }, { roles }, or { email } matching an org member. Never an address outside the org.
  • body · string · requis
    Plain-text body, max 4000 characters. Conduit wraps HTML; do not send markup.
  • title · string · requis
    Subject line, max 120 characters
  • actions · array · optionnel
    Optional buttons, max 3. Hosts outside Conduit are rejected.
  • channel · string · optionnel
    Delivery channel. Email only — omit or pass email
  • dry_run · boolean · optionnel
    If true, resolve recipients and from-address without sending or storing
  • agent_id · string · requis
    Linked agent sending the message
  • order_id · string · optionnel
    Optional order correlation
  • severity · string · optionnel · info | warning | urgent
    info (default), warning, or urgent — affects subject prefix and accent
  • search_id · string · optionnel
    Optional search correlation
  • session_token · string · optionnel
    Session token from agent_authenticate — required whenever agent_id is passed
  • idempotency_key · string · optionnel
    Replay key — same agent + key returns the original notify_id without sending again

  Exemple d'appel

  {"name": "agent_notify", "arguments": {"to": "texte", "body": "texte", "title": "texte", "agent_id": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"from": "texte", "hint": "texte", "next": {}, "error": "texte", "detail": "texte", "failed": 1, "channel": "texte", "dry_run": true, "delivered": 1, "notify_id": "texte", "recipients": [], "idempotent_replay": true}}

• agent_organization≈ 392 tokens

  Read the linked organization (name, slug, country, default address) and members (user_id, name, role, email) for an agent. Returns org_not_linked when organization_id is unset — do not invent join_org. Use members to target agent_notify.

  Paramètres, retour & exemple (2)

  Paramètres

  • agent_id · string · requis
    Agent id whose linked organization to read
  • session_token · string · optionnel
    Session token from agent_authenticate — required whenever agent_id is passed

  Exemple d'appel

  {"name": "agent_organization", "arguments": {"agent_id": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"hint": "texte", "next": {}, "error": "texte", "detail": "texte", "members": [], "connectors": [], "organization": "texte"}}

• agent_outreach≈ 865 tokens

  Email outside addresses (suppliers, carriers). to is a string email, { email }, or { emails } (max 5). Not to=all and not a Hub user_id / role — those are agent_notify. Linked agents set Reply-To to an org member (default OWNER). No action buttons. Channel is email.

  Paramètres, retour & exemple (11)

  Paramètres

  • to · ? · requis
    Outside address(es): a string email, { email }, or { emails } (max 5). Not to=all and not a Hub user_id / role — those are agent_notify.
  • body · string · requis
    Plain-text body, max 4000 characters. URLs stay escaped text, not buttons.
  • title · string · requis
    Subject line, max 120 characters
  • channel · string · optionnel
    Delivery channel. Email only — omit or pass email
  • dry_run · boolean · optionnel
    If true, resolve recipients and from-address without sending or storing
  • agent_id · string · requis
    Authenticated agent sending the note
  • order_id · string · optionnel
    Optional order correlation
  • reply_to · ? · optionnel
    Linked only: who receives replies. { email } or { role } matching a member. Omit to default to the OWNER. Unlinked agents have no Reply-To.
  • search_id · string · optionnel
    Optional search correlation
  • session_token · string · optionnel
    Session token from agent_authenticate — required whenever agent_id is passed
  • idempotency_key · string · optionnel
    Replay key — same agent + key returns the original outreach_id without sending again

  Exemple d'appel

  {"name": "agent_outreach", "arguments": {"to": "texte", "body": "texte", "title": "texte", "agent_id": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"from": "texte", "hint": "texte", "next": {}, "error": "texte", "detail": "texte", "failed": 1, "channel": "texte", "dry_run": true, "reply_to": "texte", "delivered": 1, "recipients": [], "outreach_id": "texte"}}

• agent_report_issue≈ 749 tokens

  Report unexpected tool errors or confusing Conduit outcomes for AX review (agent_report_issue — not order_feedback). Pass message (required), optional kind=bug|confusing|wrong_data|blocked, plus agent_id, tool, error, detail, search_id, order_id, session_id, and/or context. Dedupes open reports with the same tool+error+correlation. Does not change reputation.

  Paramètres, retour & exemple (11)

  Paramètres

  • kind · string · optionnel · bug | confusing | wrong_data | blocked
    Triage kind (default confusing): bug | confusing | wrong_data | blocked
  • tool · string · optionnel
    MCP tool that failed or confused you, e.g. supply_search or order_execute
  • error · string · optionnel
    Error code from the prior tool response when present, e.g. offer_not_in_cache
  • detail · string · optionnel
    Optional longer detail (response excerpt, unexpected field). Do not include private_key.
  • context · object · optionnel
    Optional structured extras (args summary, badge, etc.). Secrets are stripped.
  • message · string · requis
    Required: what went wrong or what confused you (expected vs actual). Keep actionable. This is agent_report_issue — not order_feedback.
  • agent_id · string · optionnel
    Agent id from credentials when available (omit placeholders like "agent_id")
  • order_id · string · optionnel
    order_id for correlation when the issue is order-related
  • search_id · string · optionnel
    search_id for correlation when the issue is search-related
  • session_id · string · optionnel
    Optional session id; Conduit also picks up x-conduit-session-id from the transport
  • session_token · string · optionnel
    Session token from agent_authenticate — required whenever agent_id is passed

  Exemple d'appel

  {"name": "agent_report_issue", "arguments": {"message": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"hint": "texte", "kind": "texte", "next": {}, "tool": "texte", "error": "texte", "detail": "texte", "status": "texte", "agent_id": "texte", "report_id": "texte", "session_id": "texte", "already_open": true, "reported_error": "texte"}}

• agent_update≈ 1103 tokens

  Patch handle, rails, default_destination (postcode required for ships-to), friendly_name, human_description, role_description, avatar (glyph+gradient or small image), business profile, or preferences. Set avatar only if the human chose it. Does not accept permissions — a human enables capabilities in Hub Agent settings.

  Paramètres, retour & exemple (11)

  Paramètres

  • avatar · ? · optionnel
    Identity mark. Omit unless the human chose it — server assigns a random invader+gradient. { kind:"glyph", glyph:"invader"|emoji, tint?:invader color, gradient } or { kind:"image", src }. Tint only for invader. Do not store in the credentials file.
  • handle · string · optionnel
    New display handle (lowercase kebab)
  • agent_id · string · requis
    Agent id to patch, e.g. agt_...
  • preferences · ? · optionnel
    Preferences patch
  • friendly_name · string · optionnel
    Human-facing label (≠ handle), e.g. EU Restock Bot
  • payment_rails · array · optionnel
    Replacement rail key list
  • session_token · string · optionnel
    Session token from agent_authenticate — required whenever agent_id is passed
  • business_profile · ? · optionnel
    Business profile patch
  • role_description · string · optionnel
    Self-instruction for this agent only, e.g. manage EU supplies
  • human_description · string · optionnel
    Description shown to humans on the org dashboard
  • default_destination · object · optionnel
    Where orders ship. postcode + country is enough for accurate search; add address, locality and region so checkout links arrive prefilled instead of blank.

  Exemple d'appel

  {"name": "agent_update", "arguments": {"agent_id": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"hint": "texte", "next": {}, "error": "texte", "nonce": "texte", "avatar": "texte", "detail": "texte", "handle": "texte", "persist": "texte", "agent_id": "texte", "mandates": [], "terminal": "texte", "challenge": "texte"}}

• inventory_levels≈ 577 tokens

  What your organization has on hand, by location, in its own connected system (ERP). Pass skus or query. Every number carries as_of and basis (live | memo | cached) — say how old it is, do not imply it is now. Needs inventory permission. This is your OWN stock; a supplier's stock is supply_availability.

  Paramètres, retour & exemple (6)

  Paramètres

  • skus · array · optionnel
    Item codes in your own system, e.g. ["CART-6040"]. Pass skus or query
  • fresh · boolean · optionnel
    Ask the system now instead of reading the last value (slower, rate limited)
  • query · string · optionnel
    Free-text item search in your own system, e.g. "carton"
  • agent_id · string · requis
    Agent id; its organization owns the connected system
  • connection_id · string · optionnel
    Which connected system, when the org has several
  • session_token · string · optionnel
    Session token from agent_authenticate — required whenever agent_id is passed

  Exemple d'appel

  {"name": "inventory_levels", "arguments": {"agent_id": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"hint": "texte", "next": {}, "as_of": "texte", "basis": "texte", "error": "texte", "detail": "texte", "system": "texte", "freshness": "texte", "connections": [], "availability": [], "connection_id": "texte", "could_not_reach": "texte"}}

• inventory_purchase_order≈ 518 tokens

  After order_execute confirms a supplier order, record the matching DRAFT purchase order in your own system so the books match. Always a draft: a person confirms it there. Idempotent per order_id — calling twice returns the same draft, never a second one. Needs both inventory and checkout permission. Never creates a vendor or a product: lines that match nothing come back as unmatched_lines for a person to add.

  Paramètres, retour & exemple (4)

  Paramètres

  • agent_id · string · requis
    Agent id; needs both inventory and checkout permission
  • order_id · string · requis
    Conduit order id of a supplier order already placed with order_execute
  • connection_id · string · optionnel
    Which connected system, when the org has several
  • session_token · string · optionnel
    Session token from agent_authenticate — required whenever agent_id is passed

  Exemple d'appel

  {"name": "inventory_purchase_order", "arguments": {"agent_id": "texte", "order_id": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"hint": "texte", "next": {}, "unit": "texte", "draft": true, "error": "texte", "total": 1, "detail": "texte", "vendor": "texte", "currency": "texte", "order_id": "texte", "quantity": 1, "replayed": true}}

• inventory_replenishment≈ 560 tokens

  Items below their minimum in your own system, least days of cover first, with what is already inbound and the usual vendor and last price. Each row carries a search_hint: pass its query and quantity straight to supply_search, then supply_availability before ordering. Optional location scopes to one warehouse. Needs inventory permission.

  Paramètres, retour & exemple (5)

  Paramètres

  • fresh · boolean · optionnel
    Ask the system now instead of reading the last value
  • agent_id · string · requis
    Agent id; its organization owns the connected system
  • location · string · optionnel
    Warehouse or store name to scope the worklist, e.g. "Jebel Ali". Omit for all
  • connection_id · string · optionnel
    Which connected system, when the org has several
  • session_token · string · optionnel
    Session token from agent_authenticate — required whenever agent_id is passed

  Exemple d'appel

  {"name": "inventory_replenishment", "arguments": {"agent_id": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"hint": "texte", "next": {}, "rows": [], "as_of": "texte", "basis": "texte", "error": "texte", "detail": "texte", "system": "texte", "window": "texte", "freshness": "texte", "provenance": "texte", "connections": []}}

• inventory_suppliers≈ 485 tokens

  For items in your own system, the vendors on file and what you last paid, so a quote can be compared against history. Needs inventory permission. Does not contact any vendor.

  Paramètres, retour & exemple (4)

  Paramètres

  • skus · array · requis
    Item codes in your own system whose usual vendors and last prices to read
  • agent_id · string · requis
    Agent id; its organization owns the connected system
  • connection_id · string · optionnel
    Which connected system, when the org has several
  • session_token · string · optionnel
    Session token from agent_authenticate — required whenever agent_id is passed

  Exemple d'appel

  {"name": "inventory_suppliers", "arguments": {"skus": [], "agent_id": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"hint": "texte", "next": {}, "as_of": "texte", "basis": "texte", "error": "texte", "detail": "texte", "system": "texte", "freshness": "texte", "suppliers": [], "connections": [], "connection_id": "texte"}}

• order_dispute≈ 332 tokens

  Refund/chargeback/report paths for an order. Requires owning agent_id.

  Paramètres, retour & exemple (3)

  Paramètres

  • agent_id · string · requis
    Owning agent id (required for authorization)
  • order_id · string · requis
    Order id to resolve refund/chargeback/report paths for
  • session_token · string · optionnel
    Session token from agent_authenticate — required whenever agent_id is passed

  Exemple d'appel

  {"name": "order_dispute", "arguments": {"agent_id": "texte", "order_id": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"hint": "texte", "next": {}, "error": "texte", "detail": "texte", "refund": "texte", "order_id": "texte", "chargeback": "texte", "report_to_conduit": "texte"}}

• order_events≈ 349 tokens

  Lifecycle timeline — transitions only. Pair with order_track. Requires owning agent_id.

  Paramètres, retour & exemple (3)

  Paramètres

  • agent_id · string · requis
    Owning agent id (required for authorization)
  • order_id · string · requis
    Order id whose lifecycle events to list
  • session_token · string · optionnel
    Session token from agent_authenticate — required whenever agent_id is passed

  Exemple d'appel

  {"name": "order_events", "arguments": {"agent_id": "texte", "order_id": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"hint": "texte", "next": {}, "error": "texte", "detail": "texte", "events": [], "order_id": "texte"}}

• order_execute≈ 816 tokens

  Handoff or autonomous checkout. Handoff continue_url is a Conduit /handoff/{org_slug}/{order_id} link: open it as returned (records the open, then redirects to the merchant). Autonomous needs badge payable_now (covering mandate fits). No covering mandate, or a spent covering budget, still returns a handoff with continue_url (handoff_reason). Use idempotency_key. Prefer supply_delivery when only comparing delivery.

  Paramètres, retour & exemple (9)

  Paramètres

  • carrier · string · optionnel
    The shipping rate to buy, as the option's own `title` from supply_delivery. Conduit selects it on the merchant checkout, so the buyer opens the page on that rate instead of the merchant's default, which is the cheapest one. A title the merchant does not quote for this address leaves its default in place, and the response says so via shipping_selected.
  • agent_id · string · requis
    Owning agent id
  • pay_with · string · optionnel
    Rail key to pay with when multiple are enabled
  • quantity · integer · optionnel
    Units to buy (default 1)
  • search_id · string · optionnel
    search_id from supply_search for correlation
  • supply_id · string · requis
    Offer supply id to purchase
  • session_token · string · optionnel
    Session token from agent_authenticate — required whenever agent_id is passed
  • idempotency_key · string · optionnel
    Client idempotency key — reuse to safely retry the same execute
  • payment_term_id · string · optionnel
    Payment term id from supply_details payment_terms[] (e.g. net 30). Omit to pay immediately

  Exemple d'appel

  {"name": "order_execute", "arguments": {"agent_id": "texte", "supply_id": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"hint": "texte", "next": {}, "note": "texte", "error": "texte", "detail": "texte", "status": "texte", "product": "texte", "order_id": "texte", "pay_with": "texte", "quantity": 1, "rail_key": "texte", "supplier": "texte"}}

• order_feedback≈ 583 tokens

  Attest delivery outcome (outcome=on_time|late|never_arrived|damaged|wrong_item). Requires owning agent_id. Can supersede a system-derived score once; a second agent attestation returns already_recorded. Poor outcomes next→order_dispute. Omit quality/carrier_rating if unknown.

  Paramètres, retour & exemple (8)

  Paramètres

  • notes · string · optionnel
    Free-text delivery/quality notes
  • outcome · string · requis · on_time | late | never_arrived | damaged | wrong_item
    Observed fulfillment outcome. Sets status (on_time/late→DELIVERED, damaged/wrong_item→DISPUTED, never_arrived→FAILED) and reputation.
  • quality · number · optionnel
    Product quality 1-5 — omit if unknown (do not invent)
  • agent_id · string · requis
    Owning agent id (required for authorization)
  • order_id · string · requis
    Order id to attest
  • search_id · string · optionnel
    Optional search_id for analytics correlation
  • session_token · string · optionnel
    Session token from agent_authenticate — required whenever agent_id is passed
  • carrier_rating · number · optionnel
    Carrier rating 1-5 — omit if unknown; used in reputation when provided

  Exemple d'appel

  {"name": "order_feedback", "arguments": {"outcome": "on_time", "agent_id": "texte", "order_id": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"ok": true, "hint": "texte", "next": {}, "error": "texte", "detail": "texte", "status": "texte", "outcome": "texte", "updated": "texte", "order_id": "texte", "superseded": true, "already_recorded": true}}

• order_list≈ 384 tokens

  List orders for an agent ({ orders }). Optional status or open_only filter.

  Paramètres, retour & exemple (4)

  Paramètres

  • status · string · optionnel
    Optional status filter: PENDING, CONFIRMED, IN_TRANSIT, DELIVERED, ABANDONED (or CANCELLED), FAILED, DISPUTED
  • agent_id · string · requis
    Owning agent id whose orders to list
  • open_only · boolean · optionnel
    If true, only in-flight orders (PENDING, CONFIRMED, IN_TRANSIT). Ignored when status is set.
  • session_token · string · optionnel
    Session token from agent_authenticate — required whenever agent_id is passed

  Exemple d'appel

  {"name": "order_list", "arguments": {"agent_id": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"hint": "texte", "next": {}, "error": "texte", "detail": "texte", "orders": []}}

• order_track≈ 445 tokens

  Honest deal-type-aware status (never fabricates carrier scans). Requires owning agent_id.

  Paramètres, retour & exemple (3)

  Paramètres

  • agent_id · string · requis
    Owning agent id (required for authorization)
  • order_id · string · requis
    Order id to track
  • session_token · string · optionnel
    Session token from agent_authenticate — required whenever agent_id is passed

  Exemple d'appel

  {"name": "order_track", "arguments": {"agent_id": "texte", "order_id": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"hint": "texte", "next": {}, "error": "texte", "detail": "texte", "status": "texte", "carrier": "texte", "product": "texte", "order_id": "texte", "recovery": "texte", "tracking": "texte", "trackable": true, "checkout_id": "texte"}}

• order_update_status≈ 621 tokens

  Agent manual correction or degraded handoff recovery. note required except cancelled. Owning agent_id required.

  Paramètres, retour & exemple (11)

  Paramètres

  • note · string · optionnel
    Required note except when status=CANCELLED
  • status · string · requis
    New status, e.g. PENDING, CONFIRMED, SHIPPED, DELIVERED, CANCELLED
  • carrier · string · optionnel
    Carrier name when attaching tracking
  • agent_id · string · requis
    Owning agent id (required for authorization)
  • order_id · string · requis
    Order id to update
  • search_id · string · optionnel
    Optional search_id for analytics correlation
  • tracking_id · string · optionnel
    Carrier tracking id when recovering handoff
  • tracking_url · string · optionnel
    Carrier's public tracking page for tracking_id, when you have it. Never invent one.
  • session_token · string · optionnel
    Session token from agent_authenticate — required whenever agent_id is passed
  • handoff_endpoint · string · optionnel
    Merchant handoff/status endpoint URL for poll recovery
  • external_checkout_id · string · optionnel
    Merchant checkout id to recover degraded handoff polling

  Exemple d'appel

  {"name": "order_update_status", "arguments": {"status": "texte", "agent_id": "texte", "order_id": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"hint": "texte", "next": {}, "error": "texte", "detail": "texte", "status": "texte", "carrier": "texte", "order_id": "texte", "quantity": 1, "tracking_id": "texte", "continue_url": "texte", "handoff_endpoint": "texte", "merchant_observed": "texte"}}

• payment_mandate≈ 840 tokens

  AP2 spend mandates. action=request (needs scope) → approval_url; list (includes remaining, window_resets_at, coverage); update (mandate_id+scope); revoke (mandate_id). update NEVER approves. Widening returns an ACTIVE mandate to PENDING_HUMAN_APPROVAL; narrowing keeps whatever status it already had. monthly_cap is a rolling 30-day window from create. Same category set as a live mandate replaces it at approval and keeps this period's spend. No covering mandate at checkout is handoff, not an error.

  Paramètres, retour & exemple (6)

  Paramètres

  • scope · object · optionnel
    Spend limits — required for action=request and action=update
  • action · string · requis · request | list | update | revoke
    request = create mandate (needs scope); list = enumerate; update = change scope (needs mandate_id+scope); revoke = revoke (needs mandate_id)
  • expires · string · optionnel
    Optional ISO-8601 expiry, e.g. 2026-12-31T00:00:00Z
  • agent_id · string · requis
    Owning agent id
  • mandate_id · string · optionnel
    Mandate id required for update/revoke
  • session_token · string · optionnel
    Session token from agent_authenticate — required whenever agent_id is passed

  Exemple d'appel

  {"name": "payment_mandate", "arguments": {"action": "request", "agent_id": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"hint": "texte", "next": {}, "error": "texte", "detail": "texte", "status": "texte", "applied": "texte", "mandates": [], "mandate_id": "texte", "approval_url": "texte"}}

• payment_methods≈ 428 tokens

  List or enable agent payment methods (action=list|enable) with friendly labels. Defaults: cod + x402. invoice (B2B) is coming_soon. bank_card is handoff-only.

  Paramètres, retour & exemple (4)

  Paramètres

  • rail · string · optionnel
    Rail key required when action=enable, e.g. x402 (not bank_card vault)
  • action · string · requis · list | enable
    list = enumerate rails; enable = turn on one rail
  • agent_id · string · requis
    Agent id whose methods to list or enable
  • session_token · string · optionnel
    Session token from agent_authenticate — required whenever agent_id is passed

  Exemple d'appel

  {"name": "payment_methods", "arguments": {"action": "list", "agent_id": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"hint": "texte", "next": {}, "error": "texte", "label": "texte", "detail": "texte", "status": "texte", "methods": [], "rail_key": "texte", "approval_url": "texte"}}

• supply_availability≈ 642 tokens

  Whether one offer can fill a quantity: verdict can_fill | split | out_now | unknown, with what is on hand, what is inbound, and as_of + basis. Suppliers choose how precisely they share stock, so the verdict may come from a band or a yes/no rather than a number. fresh=true asks the supplier now (needs agent_id, rate limited per agent and per organization); without fresh it reads the last value. Call this before order_execute on any connector offer.

  Paramètres, retour & exemple (6)

  Paramètres

  • fresh · boolean · optionnel
    Ask the supplier now rather than reading the last value. Needs agent_id, and is rate limited per agent and per organization
  • quote · boolean · optionnel
    Also return the supplier's exact total with tax for this quantity, computed without saving anything. Needs agent_id and an account with the supplier
  • agent_id · string · optionnel
    Agent id. Required for fresh, and it is what reveals your negotiated prices
  • quantity · integer · optionnel
    Quantity to fill (default 1)
  • supply_id · string · requis
    Offer supply id from supply_search / supply_details
  • session_token · string · optionnel
    Session token from agent_authenticate — required whenever agent_id is passed

  Exemple d'appel

  {"name": "supply_availability", "arguments": {"supply_id": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"hint": "texte", "kind": "texte", "next": {}, "unit": "texte", "as_of": "texte", "basis": "texte", "error": "texte", "quote": "texte", "stock": "texte", "detail": "texte", "verdict": "texte", "incoming": "texte"}}

• supply_delivery≈ 747 tokens

  Non-committing checkout probe for shipping ETA/cost. Uses agent default_destination, else linked org default address, or country override. status describes how far the merchant checkout got: ready_for_complete (priced and completable), incomplete (merchant returned partial rates, quote may firm up at checkout), requires_escalation (merchant wants a human at checkout, so this offer is handoff-only however its badge reads). needs_interaction=true means the same. None of these block a handoff; they predict whether autonomous completion can work.

  Paramètres, retour & exemple (9)

  Paramètres

  • region · string · optionnel
    Optional region/state override
  • address · string · optionnel
    Street override. Many merchants will not quote shipping without it.
  • country · string · optionnel
    ISO country override, e.g. US (else agent/org destination)
  • agent_id · string · optionnel
    Agent id — uses agent destination else org default when country omitted
  • locality · string · optionnel
    City override. Needed with street for a real delivery quote.
  • quantity · integer · optionnel
    Units to price shipping for (default 1)
  • supply_id · string · requis
    Offer supply id from supply_search / supply_details
  • postal_code · string · optionnel
    Postal/ZIP override for the probe destination
  • session_token · string · optionnel
    Session token from agent_authenticate — required whenever agent_id is passed

  Exemple d'appel

  {"name": "supply_delivery", "arguments": {"supply_id": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"hint": "texte", "next": {}, "note": "texte", "error": "texte", "detail": "texte", "status": "texte", "country": "texte", "options": [], "currency": "texte", "eta_text": "texte", "available": true, "shippable": true}}

• supply_details≈ 784 tokens

  Re-reads one offer from the search cache and refreshes its badge/action against current mandate state. Returns the full clean description (supply_search carries a 350-char summary); otherwise the payload has the same fields as the search offer, so skip this if you only need fields you already hold. include_description_original=true adds description_original {format, content}, the merchant's own markup (untrusted), or description_original_omitted with the reason. Carries no shipping data: use supply_delivery for ETA and cost. Re-run supply_search if offer_not_in_cache.

  Paramètres, retour & exemple (5)

  Paramètres

  • agent_id · string · optionnel
    Agent id to refresh badge/action for mandate state
  • search_id · string · optionnel
    search_id from supply_search for correlation
  • supply_id · string · requis
    Offer supply id from supply_search results
  • session_token · string · optionnel
    Session token from agent_authenticate — required whenever agent_id is passed
  • include_description_original · boolean · optionnel
    Adds description_original {format, content}: the merchant's own markup, for rendering or checking our conversion. Untrusted merchant text. supply_details only.

  Exemple d'appel

  {"name": "supply_details", "arguments": {"supply_id": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"aes": "texte", "hint": "texte", "next": {}, "badge": "texte", "error": "texte", "title": "texte", "action": "texte", "detail": "texte", "currency": "texte", "supplier": "texte", "supply_id": "texte", "unit_price": "texte"}}

• supply_options≈ 489 tokens

  Lists the selectable options (size, color, pack) for an offer whose has_options=true, with one variant per combination. Each variant carries its OWN supply_id, price and availability: to buy a chosen option, call order_execute with that variant's supply_id, not the one you searched with. Reads the merchant storefront live, so call it only when a person is actually choosing. Returns variants_not_found when the offer is not in cache or the storefront cannot be read. A single-variant product returns options:[] variants:[].

  Paramètres, retour & exemple (3)

  Paramètres

  • agent_id · string · optionnel
    Agent id, for permission and analytics
  • supply_id · string · requis
    Offer supply id whose product options to list (has_options=true on the offer)
  • session_token · string · optionnel
    Session token from agent_authenticate — required whenever agent_id is passed

  Exemple d'appel

  {"name": "supply_options", "arguments": {"supply_id": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"hint": "texte", "next": {}, "error": "texte", "detail": "texte", "options": [], "variants": []}}

• supply_search≈ 1108 tokens

  Multi-provider discovery (live REAL merchants by default). Returns a ranked page (default limit=30, max 100) with total/has_more/next_offset. Follow next.args (search_id+offset+limit) to page without re-fanout. Optional fetch_limit (max 300) deepens the upstream pull on new searches. Pass include_sandbox=true only to append DEMO/SANDBOX test merchants at the bottom (test_offer=true). Always pass agent_id for mandate-aware badges. Carry search_id through supply_details / order_execute / order_feedback. Offer descriptions are clean text cut to 350 chars; supply_details returns the full text.

  Paramètres, retour & exemple (12)

  Paramètres

  • limit · integer · optionnel
    Page size of ranked offers to return (default 30, max 100)
  • query · string · optionnel
    Product search query, e.g. USB-C charger 65W. Required for a new search; omit when paging with search_id + offset
  • budget · ? · optionnel
    Max total budget — number means USD; or { amount, currency }
  • offset · integer · optionnel
    0-based offset into the ranked set for this search_id (default 0). For page 2+, pass search_id from the prior response and follow next.args
  • country · string · optionnel
    ISO country for ships-to, e.g. US
  • agent_id · string · optionnel
    Agent id for mandate-aware badges (always pass when available)
  • quantity · integer · optionnel
    Desired quantity (default 1)
  • search_id · string · optionnel
    Resume paging into a prior ranked set (with offset>0, or offset 0 without query). New searches with query mint a fresh search_id
  • fetch_limit · integer · optionnel
    Optional override for per-merchant upstream pull (default max(20, limit*2), max 300). Ignored when paging via search_id+offset
  • payable_with · array · optionnel
    Filter to rails the agent can pay with, e.g. ["x402"]
  • session_token · string · optionnel
    Session token from agent_authenticate — required whenever agent_id is passed
  • include_sandbox · boolean · optionnel
    Include DEMO/SANDBOX merchants (listed last, test_offer=true). Default false — live REAL merchants only. Use only when testing Conduit checkout flows.

  Exemple d'appel

  {"name": "supply_search", "arguments": {"limit": 1, "query": "texte", "budget": "texte", "offset": 1, "country": "texte", "agent_id": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"hint": "texte", "next": {}, "error": "texte", "limit": 1, "total": 1, "detail": "texte", "offers": [], "offset": 1, "country": "texte", "partial": "texte", "degraded": true, "has_more": true}}

C'est votre serveur ?

Revendiquez-le sur l'espace créateur (preuve GitHub, domaine ou npm) pour afficher le badge « propriétaire vérifié » et lancer un pentest comportemental à la demande afin d'obtenir le badge « Vérifié ».

Créer un compte / revendiquer →

Source

https://mcp.aiconduit.app/mcp