agentery
$ agentery.com/api/mcpsubmit agent
Connect · MCP / REST

Give your agent live market prices.

Use Agentery inside Claude Code, Cursor and other assistants to research AI-agent markets, compare normalized pricing and monitor provider repricing. 19 tools over MCP or REST — the same pricing-intelligence layer, however you call it.

HTTP / streamable MCPREST APICurrent prices & historiesSource-backed benchmarks
claude mcp add --transport http agentery https://agentery.com/api/mcp
Add Agentery as a remote MCP server (CLI).
Quick start

Ask your assistant one question.

Once connected, ask in plain language — the assistant chains the tools for you.

you → your assistant

“What niche covers automated invoice reconciliation, what do Pro-tier providers charge, and which providers repriced recently?”

step 1find_niche

Map the task to the canonical niche.

step 2price_benchmark · niche_report

Pricing split by provider type & buyer tier + recent movement.

step 3compare_providers · get_provider_profile

Drill into specific providers where relevant.

Tools by decision

Grouped around what your agent is deciding.

The same 19 tools your agent lists over tools/list, organised by the question being answered.

01

Understand a market

What should this capability cost?

  • research_capability
  • niche_report
  • price_benchmark
  • get_price_index
  • get_price_index_history
  • market_gaps
02

Evaluate a supplier

Is this provider’s price reasonable?

  • get_provider_profile
  • compare_providers
  • get_provider
03

Find alternatives

Who can do the same job for less?

  • suggest_alternatives
  • search_providers
  • rank_providers_for_workflow
04

Monitor & signal

What’s moving and what’s missing?

  • demand_signals
  • report_outcome
  • market_gaps
Human equivalents

Core pricing outputs have a page you can read.

Not every current tool has its own human-equivalent page. The core pricing views — niches and providers — render the same numbers an agent gets, as shareable evidence.

Browse niches →Ranges · movement · what to chargeOpen a provider →Plans · percentile · history
Access

Public read, rate-limited

authentication

None required for public reads. The endpoint is open over HTTP.

read / write

15 read-only tools. The 4 writes (report_outcome + the custom-benchmark tools) only add outcome signals or manage your own private cohort — never the shared pricing data.

rate limits

Public calls are rate-limited per client; heavy or commercial use should identify itself.

freshness

Prices are observed daily; every response carries an observation timestamp.

Beyond pricing

Secondary discovery capabilities.

full docs →
discoversuggest_alternatives

Substitutes for a given agent, ranked by text-embedding nearness to its own profile (no niche lookup).

discoversearch_providers

Free-text + filtered capability search across the tracked agents and MCP servers.

discoverget_provider

Look up a single agent by handle or registration number.

discoverget_provider_profile

Full structured profile: price, pricing model, liveness, evidence URLs, how-to-connect.

discovercompare_providers

Side-by-side comparison of 2–6 providers on capability, price and evidence. Resolves handles/display-names (never silent fuzzy-swaps), returns suggested matches + a corrected_call for unknowns, and discloses mixed delivery types (agent/MCP/API).

discoverrank_providers_for_workflow

Decompose a workflow into steps and rank agent candidates per step.

Read the full API reference →