Agent Provider Depot
Server Details
Dated picks, sourced pricing, and rescue/eval tools across 305 AI agent infrastructure providers.
- Status
- Healthy
- Uptime
- 100.0% over the last 24 hours
- Last Tested
- Transport
- Streamable HTTP · MCP 2025-11-25
- URL
- ChatGPT
- –
- Claude
- –
TDQS
Scored across 4 tools
Each tool owns a distinct moment: decide for a first pick, rescue for live API errors, session_brief for change detection since a cursor, depot for catalog reads and outcome writes. The descriptions aggressively cross-route ('First pick: decide', 'Live API error: rescue'), which resolves most overlap. Minor residual ambiguity remains because depot's quote/compare/freshness ops partly echo what decide returns (sourced price, runner-up) and what session_brief reports (changes).
All names are lowercase and readable, but there is no shared pattern: decide and rescue are imperative verbs, depot is a bare noun, and session_brief is a snake_case compound. The name 'depot' is also the vaguest label for what is effectively a multi-op catalog/read tool.
Four tools map cleanly onto four workflow entry points (pick, catalog/outcome, error rescue, change brief), so each earns its place and the surface stays small. However, depot is a mega-tool compressing prices, compare, audit, freshness, map, help and report_outcome into one dispatch, which is heavy relative to the other three atomic tools.
The domain — choosing, pricing, comparing, auditing and reporting on agent providers — is well covered: first pick, quotes/compare, audit/freshness, stack-delta briefing, error rescue, and outcome reporting (the one write). Gaps are minor, e.g. no dedicated tool for fetching a single provider's integration docs or browsing the full provider list without an op dispatch through depot.
Available Tools
4 toolsdecidePick a providerARead-onlyIdempotentInspect
First pick, no API key needed, before writing integration code (LLM, sandbox, vector DB, search, email, Postgres, auth, browser, observability). First call: set a stable clientInfo.name (e.g. acme-support-agent), then {task:'run untrusted Python in a sandbox'}. Returns the dated Depot Pick, runner-up, sourced price, when to skip it, reportOutcome (decision id for depot report_outcome), recheck. Recheck: since=recheck.since (unchanged = short reply). Live API error: rescue. Stack changes since last cursor: session_brief. For quotes, compare, audit, freshness or to report an outcome: depot.
Inputs: full, jtbd, task, since, +3 more.
depotDepot drill-downAIdempotentInspect
Catalog reads and outcome reports: quotes, compare, audit, freshness. Use for those and to report an outcome. Not for a first pick (decide), a live API error (rescue), or stack changes since your last cursor (session_brief). Only report_outcome writes state; every other op is a read. Ops: prices: quote estimate compare free. Audit: check audit_repo freshness changes stack_delta coverage failures. Map: stack bundle search kit answer. Write: report_outcome. op=help lists args.
Inputs: op (required), args.
rescueRescue a failing API callARead-onlyIdempotentInspect
Live API errors: paste the exact error (429, 402, 529, insufficient_quota, model_not_found) and provider id. Returns cause, fix and fallback, or checks to run. First pick: decide. Stack changes since last cursor: session_brief. Quotes/compare/audit/freshness/outcomes: depot.
Inputs: q, sku, jtbd, status, +1 more.
session_briefEvery-session briefARead-onlyIdempotentInspect
Stack changes since your last cursor (retired models, prices, packages, picks). Send provider, model and package ids (never a manifest) plus that cursor. Returns clear|changed|attention, ≤3 sourced actions, a new cursor. First pick: decide. Live API error: rescue. Quotes/compare/audit/freshness/outcomes: depot.
Inputs: cursor, models, packages, policies, +2 more.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Related MCP Connectors
Market intelligence for the AI agent economy: rankings, trust signals, liveness. 13 tools.
Hand-verified AI infrastructure directory: tools, alternatives, EU providers, LLM API prices.
Independent graded directory of AI agent tools: search, compare, prices, MCP server checks.
Curated index of B2B offerings for AI agents: fit, non-fit, limits, sourced claims.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceCompare AI inference pricing across 9 providers in real time. Routing recommendations, spend tracking, and budget alerts for AI agents.MIT
- AlicenseNot gradedqualityNot gradedmaintenanceSearch and discover 500+ tools, APIs, and services for AI agents. Browse 15 categories, get recommendations, and access structured metadata including auth methods, free tiers, and example calls.1-
- FlicenseNot gradedqualityDmaintenanceEnables real-time access to LLM pricing, benchmarks, deprecation alerts, and cost optimization for over 30 models across 8 providers, allowing AI agents to make cost-effective model selections.-
- FlicenseNot gradedqualityBmaintenanceIntelligently routes AI tool calls (search, extract, browse) to the cheapest capable provider by default with automatic failover and audit receipts, enabling cost-effective agent workflows.2-
Glama MCP Gateway
Add one secure layer between your agents and this server.