sellable-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| search_catalogA | Search products by text, optionally filtered by category and maximum price in rupees. Prices are server-side and authoritative. |
| get_productB | Full record for one SKU: price, attributes, compatibility, policies, stock. |
| list_categoriesA | All product categories with item counts. |
| list_missionsA | Missions the user has signed: intent, budget, remaining budget, allowed and forbidden categories, expiry. You can only buy inside one of these. |
| propose_purchaseA | Ask the deterministic gateway to approve a basket for a mission. You send only SKUs and quantities. On APPROVE you get a single-use approval_id valid for 10 minutes; on REJECT you get the rule that refused it and a hint. |
| place_orderA | Execute a basket the gateway approved. Takes only the approval_id; the amount is the catalog total of the approved basket. Each approval works once. |
| get_orderC | Look up an order by id. |
| list_ordersC | Orders placed so far, optionally for one mission. |
| get_policyC | The deterministic rules that gate every purchase. |
| verify_audit_logB | Recompute the hash chain over every proposal, verdict and order. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 10 tools
Each tool targets a distinct stage or resource in the purchase flow: catalog search vs. single product lookup vs. category listing, and the propose_purchase/place_order split cleanly separates approval from execution. No two tools overlap in purpose, so an agent can select correctly with confidence.
Uniform verb_noun snake_case throughout (search_catalog, get_product, list_missions, place_order, verify_audit_log, etc.). The convention is predictable and consistent with no mixed styles.
Ten tools is well-scoped for a constrained purchasing agent, covering discovery, approval, execution, and audit. Every tool earns its place without redundancy.
The surface covers the full purchase lifecycle (search, inspect, mission constraints, propose, order, lookup, audit) with no dead ends in the core flow. Minor gaps exist—no order cancellation or mission management—but these may be intentionally out of scope since missions are externally signed.