Robin Saige we check the tools AI agents call

nullary

ai.nullary/nullary

Negative results intelligence for drug discovery — query measured failures via MCP.

— the operator's own registry description reported

Publisher ai.nullary · first seen 2026-08-12 · endpoint https://mcp.nullary.ai/mcp

allow
verdict
rich-solo
cluster
ok_tools
answers
35
tools served

Dependency rating derived

allow — rich-solo — safe to depend on

Cluster rich-solo · distinctiveness 0.378 (moderate). Every figure is a percentile or class within the 11,094-server censussee the whole spectrum.

Reach — can an agent get to it?
answersalive_tools
latency103 ms
vs populationtypical
protocolcurrent
authopen
Use — can it use it?
tools35
vs populationhigh
capabilities1
harnessA
Trust — can it rely on it?
identitysolo
servers on its host1
verifiabilityunassessed
duplicate inventoryno
drift events0
answers trueno primary

🔏 Signed receipt rr_3ec7d559d54de75c042b4bd5 · 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.

Tools last seen

get_compound get_coverage get_finding_provenance get_model_card get_target_landscape list_models list_top_targets search_adc_linker_failures search_admet_failures search_admet_failures_all_modalities search_ancestry_specific_failures search_bispecific_format_failures search_developability_failures search_drug_drug_interaction_failures search_failed_adcs search_failed_bispecifics search_failed_clinical_antibodies search_failed_essentiality_screens search_failed_guides search_failed_oligonucleotides search_failed_peptide_therapeutics search_failed_protacs search_failed_replications search_failed_selectivity search_failed_vaccines search_inactive_compounds search_indication_history search_mechanism_failures search_oligo_delivery_failures search_pathogen_history search_peptide_stability_issues search_protac_e3_issues search_safety_failures search_target_history search_vaccine_immunogenicity_failures

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 & specific35/35
has a description35/35
description is substantive21/3521/35
parameters are typed35/35
parameters are described35/35

35 of 35 tools graded · 35 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
1.0.1
server version Nullary

Capabilities: tools

Drift

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

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
raw probe records
probed (UTC) outcomehttpmsprotocol
2026-08-14T19:57:37Zok_tools2001032025-06-18
2026-08-12T05:32:55Zok_tools2001782025-06-18

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/ai.nullary/nullary.svg)](https://robinsaige.com/s/ai.nullary/nullary)

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