Robin Saige we check the tools AI agents call

Glasswarp

com.glasswarp/mcp-server

Eyes and hands on real Windows PCs — observe, click, type via Glasswarp API.

— the operator's own registry description reported

Publisher com.glasswarp · first seen 2026-08-12 · endpoint https://mcp.glasswarp.com/mcp

allow
verdict
mid-solo
cluster
ok_tools
answers
16
tools served

Dependency rating derived

allow
allow — alive & usable · action ● · V n/a
allow — because answers ●, 16 tools legible (band A) ●, no confirmed drift in 2 probes ●; not yet checked: truth (0 runs)

Cluster mid-solo · distinctiveness 0.471 (moderate). Every figure is a percentile or class within the 11,094-server censussee the whole spectrum.

Does it answer? Q1 · observed
answersalive_tools
latency76 ms
vs populationtypical
protocolcurrent
authopen
Can it be used? Q4 · observed
tools16
vs populationmid
capabilities3
harnessA
Did it hold when checked? Q3–Q4 · derived
identitysolo
servers on its host1
verifiabilityunassessed
duplicate inventoryno
drift events0
answers trueno primary

🔏 Signed receipt rr_213f87cbd16cca81eddc01f7 · 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?action ● evidence ↓
Q3Can it be checked?V n/a for this kind ◐evidence ↓
Q4Did it hold?harness A · truth not yet run · no drift ●evidence ↓
Q5What company does it keep?mid-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
action10changes the worldno true answer exists; we check the CONTRACT (destructive/read-only/idempotent declarations, error legibility) and never fire real actions
retrieval-public6serves public factstruth-checkable against a public source — the V-ladder applies in full

Tools last seen observed

app.launch demos.get demos.list input.click_target input.click_xy input.drag input.scroll input.send_actions input.send_keys input.type_text rigs.list screen.observe session.end session.live_view session.start session.status

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 & specific16/16
has a description16/16
description is substantive16/16
parameters are typed16/16
parameters are described16/16

16 of 16 tools graded · 16 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 in the Tier-2 trade lane, or not yet checked. Tier-2 re-derives a server's answers against published primary sources — see truth checks.

Protocol conformance observed

2025-06-18
protocol version
session model
0.1.0
server version glasswarp

Capabilities: prompts 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
2026-08-14T19:58:10Zok_tools200762025-06-18
2026-08-12T05:33:25Zok_tools200842025-06-18

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/com.glasswarp/mcp-server.svg)](https://robinsaige.com/s/com.glasswarp/mcp-server)

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