price benchmark

Fair-price benchmarks for a capability described in natural language. The MARKET is the semantic neighbourhood of your query — the nearest providers by text emb

liveverifiedunknown~5s typical

mcpprobedagentery

$0.0020

per call

plus $0.001 platform fee per run

Open MCP server; Cracked bills routing only

Provider Upstream docs

$Set up https://cracked-ai-pearl.vercel.app/SKILL.md, then use Cracked to run mcp-com-agentery-agentery /price_benchmark for me.

Input

Field	Type	Description
niche	string	Legacy alias of `query`, kept for older clients — prefer `query`.
sector	string	Sector name, e.g. 'legal' (ignored if a query is given)
provider_type	string (provider \| mcp \| api \| all)	Delivery type being priced. Set when the user says provider, MCP or API. Omit (or 'all') t
buyer_tier	string (individual \| pro \| team_sme \| enterprise \| all)	Buyer tier being priced. Set when the user describes who is buying (an individual, a profe
pricing_unit	string	Optional pricing unit to hold constant (e.g. 'flat', 'per_seat', 'per_agent'). Incompatibl
response_mode	string (summary \| full)	'summary' (default): compact per-type/per-tier benchmark matrix. 'full': also returns the

Call it

curl

curl https://cracked-ai-pearl.vercel.app/v1/run \
  -H "Authorization: Bearer ck_live_..." -H "content-type: application/json" \
  -d '{"provider":"mcp-com-agentery-agentery","endpoint":"/price_benchmark","input":{}}'

cli

npx cracked-ai run -p mcp-com-agentery-agentery -e /price_benchmark -i '{}'

mcp

run_tool({ provider: "mcp-com-agentery-agentery", endpoint: "/price_benchmark", input: {} })

Try it

Runs against your signed-in workspace balance. Sign in if you have not.