SociaBench bronze

www.sociabench.com

“Read-only data from SociaBench (AI Social Intelligence Leaderboards): benchmarks, leaderboards, model results, reproduction reports and evaluation protocols. The leaderboards include only evaluations this platform ran under the current protocol version. Results published by the authors are listed separately in the original leaderboard, do not count toward the ranking, and cannot be compared with it directly. When you cite, credit "SociaBench (AI Social Intelligence Leaderboards)" with the link to the page, and give the formal name and original source of each benchmark. Text is American English by default; the tools that return page or benchmark text take a `language` argument (en-US or zh-CN). This agent does not use a language model: a request names a skill and its parameters in a data part, or gives a benchmark ID, benchmark name or page path as text.

a2a https://sociabench.com/a2a talk to it https://www.sociabench.com/.well-known/agent-card.json its card

Checked 3h ago pushNotifications, stateTransitionHistory, streaming

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 Google Trust Services domain-validated
Valid until 25 Jan 2027 as of our check on 15 Nov 2026.

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

Key ECDSA P-256, signed ECDSA-SHA256, issued via WE1 under GTS Root R4 — classical, as every publicly trusted certificate is today.

Names on this certificate (1)

sociabench.com.
Connection: TLS 1.3 · X25519MLKEM768 — post-quantum hybrid key exchange, agreed with our checker on 15 Nov 2026. What this means
DANE / TLSA: Not verified
Discovery: Well-known document
AI use policy: Search: yesAI input: yesAI training: yes

What this site's robots.txt says about how AI may use its content. Recorded as the operator wrote it, not enforced — these are preferences about use, not access, and agenttru.st only reads the agent's own discovery documents.
AI-facing documents: Publishes /llms.txt — “SociaBench (AI Social Intelligence Leaderboards)” 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: 15 Nov 2026

View verification details

Assurance: bronze Bronze — agent card fetched over HTTPS with a valid certificate
Protocols: A2A verified by handshake or card fetch, not merely advertised
Hosted in: ? Unknown · Cloudflare, Inc. (AS13335)
The address did not geolocate — usually anycast hosting, where one address answers from many places at once.
Last checked: 3h 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.sociabench.com controls every word below.

List benchmarks

Lists every benchmark SociaBench covers: ID, name, category, summary, original source, and the addresses of its leaderboard and reproduction report.

SociaBenchAI social intelligenceleaderboards

Examples it gives

{"skill": "list_benchmarks"}
list benchmarks

Get leaderboard

Reads a benchmark's leaderboard (live data): the platform's ranking (overall score for every view, 95% confidence interval, group scores, sample count, evaluation time and data version), plus the original leaderboard of results published by the authors (listed separately, not part of the ranking).

SociaBenchAI social intelligenceleaderboards

Examples it gives

{"skill": "get_leaderboard", "benchmark": "characterbench"}
{"skill": "get_leaderboard", "benchmark": "characterbench", "language": "zh-CN"}
characterbench

Get model result

Reads a model's platform evaluation result on a benchmark (live data): its rank in each view, overall score and confidence interval, group and by-dimension scores, model version and evaluation time. Only models this platform evaluated have a result page; models that appear only in the original leaderboard do not.

SociaBenchAI social intelligenceleaderboards

Examples it gives

{"skill": "get_model_result", "benchmark": "characterbench", "model": "deepseek-v4-1-flash-261017"}

Get page

Reads the Markdown version of any public SociaBench page, for example the reproduction report /b/characterbench/methodology, the research list /methodology or the Data Platform Terms /data-terms.

SociaBenchAI social intelligencereproduction report

Examples it gives

{"skill": "get_page", "path": "/b/characterbench/methodology"}
/b/coser/methodology
/zh-CN/b/coser/methodology

Get evaluation protocol

Reads a benchmark's full evaluation protocol on this platform (JSON): protocol version, the prompts of the model under test and the judge, judge configuration, scoring rules and run parameters, for checking and reproduction.

SociaBenchAI social intelligenceevaluation protocol

Examples it gives

{"skill": "get_protocol", "benchmark": "minimax-rp"}

Technical agent card

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

Provider: SociaBench — what this agent says about itself; other agents claiming the same provider are not thereby related
Protocol: a2a
Version: 1.0.0
Card completeness: complete all eight fields required by a2a.proto v1.0

View all card details

Capabilities: pushNotifications stateTransitionHistory streaming
Agent card: https://www.sociabench.com/.well-known/agent-card.json

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