AP
Agent Provider Depot
Opinionated picks for agents
OpenAPI
← All jobs

Hosted MCP server runtime

Deploy and scale MCP servers (tools/resources) with auth, logging, and HTTPS without babysitting containers — discovery alone is not enough.

At a glance — Oct 2026

Updated 2026-10-27 · forward this page: /jtbds/mcp-server-hosting#brief

Scores as of 2026-09-14 (43 days old; Depot marks scores stale after 30 days). The pick is still Depot's current editorial pick; it has not been re-reviewed since. Prices below carry their own dates.

Depot Pick
Cloudflare Workers · 8.52/10 for this job (curated estimate, high confidence)

Distinct from `mcp-discovery`. Editorial replacement 2026-10-24: Smithery's own page (https://smithery.ai/future, fetched 2026-10-24) says 'Server URLs on run.tools and server.smithery.ai' stop working on October 31, 2026, so the former pick smithery (7.87 for this job) was replaced by the highest-scoring live runner-up, Cloudflare Workers (8.52 for this job), and removed from this job. No score changed. Confidence: medium — managed hosting SKUs move.

Runner-up
Glama · 7.65/10

Scores higher than the pick on: Lock-in risk (higher = worse).

When not to use the pick
Only need registry listing (mcp-registry/Glama), or Composio-managed tools replace custom MCP.
Lock-in / migration
Runner-up Glama has lower stored lock-in risk (4 vs 5.5, 0–10, higher = more lock-in).
Sourced price (no comparable unit)

Pick: no sourced price in this job's unit (no_matching_sourced_price).

No cheapest SKU (no_comparable_unit_for_job). No price was invented.

Free option
No free resource lists this job. A cheap paid SKU is not a free option.
Failure modes to watch
  • Hallucinated tool arguments: Reject a call that fails the current schema or names an unknown id, and do that before the tool runs a side effect.
  • Schema drift: Validate every tool call and every loaded table against the current schema. Quarantine mismatches instead of coercing them.
  • Tool loop: Hash the last several tool names and arguments. Stop when the same call repeats and the observation did not change. Cap tool calls per task.
Reported outcomes (submitted reports only)

This job: loading… JSON · reports never change scores · seed/test records excluded

Depot Pick (Cloudflare Workers) on this job: loading… JSON · reports never change scores · seed/test records excluded

Cite: Agent Provider Depot, Depot Pick for mcp-server-hosting: cloudflare-workers, scores as of 2026-09-14. Agents: GET /api/decide?jtbd=mcp-server-hosting

Depot Pick

Cloudflare Workers

Distinct from `mcp-discovery`. Editorial replacement 2026-10-24: Smithery's own page (https://smithery.ai/future, fetched 2026-10-24) says 'Server URLs on run.tools and server.smithery.ai' stop working on October 31, 2026, so the former pick smithery (7.87 for this job) was replaced by the highest-scoring live runner-up, Cloudflare Workers (8.52 for this job), and removed from this job. No score changed. Confidence: medium — managed hosting SKUs move.

Pricing (directional): Request-based Workers + Durable Objects + R2 + Queues; generous free tier; CPU time limits.

Cite: Depot Pick for mcp-server-hosting: cloudflare-workers
Overall
8.5
curated_estimate
confidence: high · as of 2026-09-14
Capability fit
8.6
Price / performance
9.2
Latency / reliability
9.2
Developer & agent DX
8.5
Privacy / data retention
8.0
Lock-in risk (higher = worse)
5.5
Maturity / track record
9.5

Lock-in risk: higher = more lock-in (worse). Overall already inverts this dimension.

Sources: Cloudflare Workers: https://workers.cloudflare.com · Cloudflare AI: https://developers.cloudflare.com/workers-ai/

Best edge runtime for lightweight agent gateways and tools. Heavy LLM loops still belong on Modal/Fly/GPU hosts.

Quote for Hosted MCP server runtime

Depot Pick stays Cloudflare Workers. Prices below are sourced SKU rows only (as of 2026-10-27). Input and output are not blended. GET /api/quote?jtbd=mcp-server-hosting

Nothing was compared. no comparable sourced unit for this job; chat input $/1M and other units were not borrowed pickIsPriceWinner is null.

  • D1 rows read (Workers Paid) usd_per_million_rows_read $0.001 source as of 2026-10-21
  • D1 rows written (Workers Paid) usd_per_million_rows_written $1 source as of 2026-10-21
  • D1 storage (Workers Paid) usd_per_gb_month $0.75 source as of 2026-10-21

2 of 4 candidates have no sourced price row. 4 are excluded from the Matching price comparison (none). no comparable sourced unit for this job; chat input $/1M and other units were not borrowed

Runners-up

When NOT to use the top pick

Only need registry listing (mcp-registry/Glama), or Composio-managed tools replace custom MCP.

Full ranking (score table)

Ranked with this JTBD's weights. Scores are curated estimates unless marked otherwise. Agents: prefer /api/decide then this table for tie-breaks.

#ProviderOverallFit$/perfLatencyDXAs of
1Cloudflare WorkersDepot Pick8.58.69.29.28.52026-09-14
2Fly.io8.48.88.28.08.82026-09-14
3Composio7.88.27.57.88.82026-10-19
4Glama7.77.88.57.28.22026-09-14

Score methodology for this job

Distinct from `mcp-discovery`. Confidence: medium — managed hosting SKUs move; Smithery's hosted server URLs end 2026-10-31 (pick replaced 2026-10-24).

  • capabilityFit: 22%
  • pricePerformance: 14%
  • latencyReliability: 14%
  • developerDx: 28%
  • privacyPosture: 8%
  • lockInRisk: 6%
  • maturity: 8%