Undercut public data bronze

www.getundercut.sh

“Read-only access to data Undercut (https://getundercut.sh) already publishes in the undercutsh/firstpass repository. Every tool returns the same JSON as the matching GET /api/* endpoint (described at https://getundercut.sh/openapi.json), as structuredContent plus a JSON text block. Nothing here routes, proxies, or runs model calls, and there are no write tools: the routing itself happens inside your own agent, which follows the policy that get_policy returns. Benchmark claims should cite get_benchmark_results, including its standingLimitation and caveats.

https://getundercut.sh/mcp its card

🇺🇸 US · Amazon.com, Inc. Checked 5d ago tools

we checked this the operator says this

Community rating 0 0 up · 0 down — sign in to vote

Verified by agenttru.st

Everything here is a check agenttru.st performed itself. Assurance, protocol, hosting and freshness are in the card above and are not repeated.

Certificate: Issued by Let's Encrypt domain-validated
Valid until 17 Jan 2027 as of our check on 9 Nov 2026.

Control of the hostname was checked; nothing about who operates it.

Key RSA 2048, signed SHA256-RSA, issued via YR1 under ISRG Root X1 — classical, as every publicly trusted certificate is today.

Names on this certificate (1)

getundercut.sh.
Connection: TLS 1.3 · X25519MLKEM768 — post-quantum hybrid key exchange, agreed with our checker on 9 Nov 2026. What this means
DANE / TLSA: Not verified (TLSA query returned RCodeNameError)
Discovery: Well-known document
This host also publishes an ARD catalog (/.well-known/ai-catalog.json) declaring 6 resources. The catalog's entries are the publisher's claims, not something agenttru.st verified.
AI-facing documents: Publishes /llms.txt — “llms.txt — Undercut” a curated map of the site's content for language models

Fetched from this host during verification. Neither document is how this agent was discovered.
First seen: 2 Nov 2026

View verification details

Assurance: bronze Bronze — agent card fetched over HTTPS with a valid certificate
Protocols: MCP verified by handshake or card fetch, not merely advertised
Hosted in: 🇺🇸 US · Amazon.com, Inc. (AS16509)
Last checked: 5d ago

What this agent says it can do

Declared in the agent's own card. agenttru.st has not tested whether it completes any of these tasks — the operator of www.getundercut.sh controls every word below.

This card declares no skills. That is not a fault — the field is optional, and an agent can be perfectly functional without describing itself.

What kind of listing

No skills declared: Its card declares no skills, or, for an MCP server, listed no tools.

Checked by agenttru.st against this listing's hostname and card — facts about a name or an address, not about what the agent does. How kinds are decided.

What its catalog declares

From the ARD catalog this host publishes at /.well-known/ai-catalog.json. The operator controls every entry; agenttru.st verified that the catalog exists and parses, and nothing about what the entries say.

Undercut — tiered dispatch

text/markdowntrust manifest

A six-flag rubric and escalation ladder that a coding agent follows at dispatch time to route each unit of work to the cheapest model tier that passes objective verification. Free, MIT licensed. Install with `npx skills add undercutsh/firstpass`.

https://raw.githubusercontent.com/undercutsh/firstpass/main/skills/firstpass/SKILL.md

undercutsh/firstpass (source)

text/htmltrust manifest

Source repository: the skill, the vendor-constant deterministic eval harness (evals/), and the raw public benchmark results (testing/).

https://github.com/undercutsh/firstpass

llms.txt

text/plaintrust manifest

Agent-facing navigation index: what Undercut is, when to use it, benchmark summary, and links to docs.

https://getundercut.sh/llms.txt

Undercut public data API (OpenAPI 3.1)

vnd.oai.openapitrust manifest

OpenAPI 3.1 description of Undercut's read-only public JSON API. Every operation is an unauthenticated GET; there are no write endpoints and no auth flow.

https://getundercut.sh/openapi.json

Undercut public data API

jsontrust manifest

Read-only, unauthenticated JSON over data Undercut already publishes: the routing policy (/api/policy), the tier-to-model mapping (/api/models), benchmark results computed from the raw files in testing/results (/api/results), supported clients (/api/clients, /api/clients/{slug}), the setup segment router (/api/segments), pricing (/api/pricing), and published Teams onboarding windows (/api/teams-availability). CORS is open. It does not route, proxy, or run model calls.

https://getundercut.sh/api/health

Undercut MCP server (read-only)

mcp-server-cardtrust manifest

Server Card for Undercut's read-only, unauthenticated MCP server at https://getundercut.sh/mcp (Streamable HTTP, stateless, JSON responses; protocol 2026-07-28 plus 2025-11-25, 2025-06-18 and 2025-03-26). Tools: list_clients, get_client, get_segments, get_policy, get_models, get_benchmark_results, get_pricing, get_teams_availability. Each returns the same published data as the matching /api/* endpoint. No write tools, no auth, no routing or inference.

https://getundercut.sh/mcp/server-card

Technical agent card

Copied from the agent's card. The operator controls these values; agenttru.st has not verified them.

Protocol: mcp
Version: 0.5.0

View all card details

Capabilities: tools
Agent card: https://getundercut.sh/mcp

Operate this agent and would rather not be listed? Request removal.