Robin Saige we check the tools AI agents call

Curie Commerce

co.curie/commerce

Search, compare in 3D, and buy products from 5.6M+ Shopify stores. Free tier; Pro adds checkout.

— the operator's own registry description reported

Publisher co.curie · first seen 2026-08-12 · endpoint https://chat.curie.app/api/mcp/global

allow
verdict
mid-slow-solo
cluster
ok_tools
answers
30
tools served

Dependency rating derived

allow
allow — alive & usable · retrieval-public ● · V3 has answer key ◐
allow — because answers ●, 30 tools legible (band A) ●, no confirmed drift in 2 probes ●; not yet checked: truth (3 attempts unverifiable, 0 disagreements)

Cluster mid-slow-solo · distinctiveness 0.641 (unusual). Every figure is a percentile or class within the 11,094-server censussee the whole spectrum.

Does it answer? Q1 · observed
answersalive_tools
latency680 ms
vs populationslow
protocolcurrent
authopen
Can it be used? Q4 · observed
tools30
vs populationmid
capabilities3
harnessA
Did it hold when checked? Q3–Q4 · derived
identitysolo
servers on its host1
verifiabilityunassessed
duplicate inventoryno
drift events0
answers true0/0 agree

🔏 Signed receipt rr_37dbaaa5c82befe445aa5625 · ed25519 · key rs-rcpt-2026-08 — this rating is tamper-evident; an agent gets the full signature from check_server and can verify it against the published key.

Q1Does it answer?alive & usable — 2 probes ●evidence ↓
Q2What is it?retrieval-public ● evidence ↓
Q3Can it be checked?V3 has answer key ◐evidence ↓
Q4Did it hold?harness A · truth not yet run · no drift ●evidence ↓
Q5What company does it keep?mid-slow-solo · 1 on its host (context, not verdict) ◐evidence ↓

● observed · ◐ derived · ○ reported — the verdict synthesizes Q1–Q4; Q5 is context. Methodology →

History observed

Every probe we made, oldest → newest (2 shown, 2026-08-12 → 2026-08-14). Green answered · amber answered-but-walled · red no useful answer. A gap in our cadence is a gap in coverage, not evidence about the server.

changelog (1 events)
datetypewhat changed
2026-08-12outcomefirst probe: ok_tools

What kind of tools these are derived

Verification only means something relative to a tool's NATURE — an email-sender has no "true answer", a generator has no answer key. Classification evidence: classified from tool names/descriptions/annotations WE OBSERVED. What checking applies to each kind (the full methodology →):

naturetoolsmeaning what checking applies
retrieval-public15serves public factstruth-checkable against a public source — the V-ladder applies in full
action14changes the worldno true answer exists; we check the CONTRACT (destructive/read-only/idempotent declarations, error legibility) and never fire real actions
retrieval-private1serves the caller's own dataoutside truth-checking is impossible BY NATURE, not by failure — the operator-invited pathway applies

Tools last seen observed

add_product_from_photo cancel_cart cancel_checkout compare_products complete_checkout create_cart curie_cart_aggregate_add curie_cart_aggregate_get curie_cart_aggregate_remove curie_cart_aggregate_update_quantity curie_try_on_3d curie_wishlist_add curie_wishlist_get curie_wishlist_move_to_cart curie_wishlist_remove get_availability get_brand_context get_cart get_checkout get_order get_product_details get_recommendations get_suggested_prompts initiate_checkout lookup_catalog search_catalog search_shop_policies_and_faqs solve_job update_cart update_checkout

Harness readiness observed

Can an agent's harness pick this tool and call it correctly? Graded on the three things an agent reads — name, description, typed parameters. Band A — an agent can pick and call these reliably.

legibility checktools passingok
name is clear & specific30/30
has a description30/30
description is substantive30/30
parameters are typed30/30
parameters are described30/30

30 of 30 tools graded · 30 with a captured input schema. RS-008, Tier-1 — no domain knowledge, applies to any tool. “Alive” is not the same as “usable”.

Truth checks observed

Not yet confirmed — 3 attempt(s) currently unverifiable (rate limits, walls, or unmappable schemas — never counted against the server) · 0 disagreements.

claim classkeyverdictprimary source
tariff-duty0901.21.00unverifiableus-hts
tariff-duty6109.10.00unverifiableus-hts
tariff-duty8471.30.01unverifiableus-hts

Protocol conformance observed

2025-06-18
protocol version
session model
1.0.0
server version curie-commerce-global

Capabilities: experimental resources tools

Drift observed

No confirmed drift on record. Baseline set 2026-08-14; changes appear here after human review.

raw probe records
probed (UTC) outcomehttpmsprotocol
2026-08-14T19:57:51Zok_tools2006802025-06-18
2026-08-12T05:33:08Zok_tools2006762025-06-18

Watch or embed this verdict

verdict badge

Operators: put the live verdict in your README — it updates with every census, links back here, and is a dated observation, never a warranty:

[![Robin Saige verdict](https://robinsaige.com/badge/co.curie/commerce.svg)](https://robinsaige.com/s/co.curie/commerce)

Depend on it? Subscribe to its change feed — outcome changes and confirmed drift, no account needed. Building on it? The full dossier as JSON — verdict, rating, truth checks, history — stable enough to gate CI on.

← overview