{
  "data": null,
  "kind": "anchor.page",
  "links": {
    "api": "https://www.anchorterminal.com/api/v1/index.json",
    "html": "https://www.anchorterminal.com/niches/decision-models.gate-action/",
    "json": "https://www.anchorterminal.com/niches/decision-models.gate-action/index.json",
    "llms": "https://www.anchorterminal.com/llms.txt",
    "markdown": "https://www.anchorterminal.com/niches/decision-models.gate-action/index.md",
    "mcp": "https://www.anchorterminal.com/mcp",
    "slim": "https://www.anchorterminal.com/niches/decision-models.gate-action/index.min.md"
  },
  "markdown": "Four or more companies' docs show a full fit.\n\n| Companies, full fit | With a limit | Carry the tag | Share of the category's work | First-hand accounts |\n| --- | --- | --- | --- | --- |\n| 9 | 12 | 42 | 16 | 8 |\n\n## What counts as done\n\nA probability of yes that can be thresholded.\n\nMeasured, for our own tests: at the chosen threshold no irreversible action proceeds on a wrong yes in the caller's labelled set, under 300 ms.\n\nHow agents ask for it: is this tool call safe to run; does this message authorise the booking; is this request a prompt injection; does this line affect driving safety\n\n## Does the job\n\nThe docs show it meets the test above, and nothing we read says otherwise.\n\n- **Celeris-1 Decision (Marqo Inc)**: noul returns the probability of yes; the Policy gates section says to ask whether the input violates the policy and pick thresholds from labelled examples. 131,072-token context, so a long policy fits.\n- **Clef (Cloudflare, Inc.)**: noul returns the probability of yes; 65,536-token context on Clef. Arize found Clef kept 0.92 and 0.91 AUC with the question worded both ways, one of three models that did. No prompt-injection guidance from Cloudflare.\n- **elephant-1 (Baseten Labs, Inc.)**: noul returns P(yes); trained at 16,384 tokens with longer requests refused, a trained 'I don't know' option and 16 percent safety and abstention data. No prompt-injection guidance.\n- **joey-1 (Baseten Labs, Inc.)**: noul returns P(yes); trained at 16,384 tokens with longer requests refused, a trained 'I don't know' option and 16 percent safety and abstention data. No prompt-injection guidance.\n- **mammoth-1 (Baseten)**: noul returns P(yes); trained at 16,384 tokens with longer requests refused, a trained 'I don't know' option and 16 percent safety and abstention data. No prompt-injection guidance.\n- **Drex 1.5 (Nace AI, Inc.)**: noul with true and false criteria; drex-v2.0 reads states to 131,072 tokens and refuses longer ones with a 422 naming the limit, so a long policy need not be compressed.\n- **OpenAI Decisions API (OpenAI OpCo, LLC)**: predicate returns P(true); input can carry conversation history, tool calls and tool outputs; thresholds from labelled examples; 922,000 input tokens on GPT-6 Luna. Arize found it kept 0.90 and 0.89 AUC with the question reversed. No adversarial-input guidance.\n- **pplx-decider (Perplexity AI, Inc.)**: noul with criteria, 256K context, and published calibration (ECE 0.014, Brier 0.134 on the development set; about 0.05 on held-out decision datasets). No injection guidance.\n- **pplx-decider-flash (Perplexity AI, Inc.)**: noul with criteria, 256K context, calibration published (ECE 0.021, Brier 0.163 on the development set).\n- **kettei (Sqwish Ltd, trading as Sqwish Labs)**: binary decisions with abstain_below (examples at 0.4 to 0.7) and a tool-call check example; 65,536-token input limit with a 422 context_too_long rather than a cut.\n- **Jev (TypeSafe AI, Inc.)**: noul with 0.7 and 0.3 as starting thresholds and a worked risk-tiered example (0.5 floor, 0.9 for a transfer); OpenRouter's screen caught 40 of 40 injections; the jaggedness page says adversarial content can steer answers; 64k context, 262k by header.\n\n## Does the job, with a limit\n\nIt does the job, but something in the docs or a user's account will get in an agent's way. The limit is named in each row.\n\n- **Decider (Mark Marosi (Mapika))**: noul with per-type temperatures, but the README says rules written into the question are not followed at this size (0.24 against 0.67) and calibration on hard items is weak; the maintainer advised acting on destructive actions only at 0.8 or above.\n- **Decision 3.0 (vLLM Semantic Router project)**: noul with a probability of yes and no truncation, but issue 6298 found the vela3 and d3 sizes used as jailbreak and harm guards missed attacks the base model refused.\n- **Drex Max v1 (Nace AI, Inc.)**: noul with a 16,384-token default raisable to 262,144 (DREX_MAX_CONTEXT). No safety results, no prompt-injection guidance and the licence cap.\n- **Granite Switch 4.1 (International Business Machines Corporation)**: A guardian-core safety adapter and a decision adapter share the endpoint, with the 128K base context for a long policy; there is no typed noul, so the gate is a two-label logprob read.\n- **Kev (Jared Palmer)**: noul with calibrated probabilities and a 65,536-token server limit, but Arize found Kev 9B's AUC fall from 0.85 to 0.54 when the question was reversed, and the small models are validated to 8,192 tokens.\n- **JEV-27B-VL (AutoTrust AI Lab)**: Yes/no decisions over up to 256K tokens of text (20 of 20 at every depth tested), one question a request, with no safety results or injection guidance.\n- **Liquid d1 (Liquid AI, Inc.)**: noul with a worked binary-decision example in the guide, but Arize saw d1's AUC fall from 0.92 to 0.17 when the question was reversed, so the gate must be tested both ways.\n- **Microsoft-Decision-1 (Microsoft)**: noul with criteria for true and false, 32,768-token context and a vendor safety test over 5,250 requests; no refusal type, and 0.42 s on OpenRouter is over the 300 ms target.\n- **RightNow System Zero (RightNow AI, Inc.)**: boolean returns the probability of true with a refusal per question, over 38,000 state tokens, but the probabilities are flagged uncalibrated, so a threshold has no stated meaning.\n- **ScaleDown (Carbon ScaleDown, Inc.)**: noul returns an independent P(yes); no context limit, calibration figure or refusal type is documented, and no key without sales contact.\n- **SPEV (Sprag AI, Inc.)**: predicate returns P(true) with a refusal type and a 400 model_limit_exceeded naming the limit; the weights' config caps input at 8,192 tokens, short for a long fixed policy.\n- **Strands Decider 2B (Amazon Web Services, Inc.)**: noul works, but at a 0.95 threshold no booking proceeded and at 0.50 two unauthorised ones passed; injection recall 92.5 percent at 17 percent false positives, weak on encoded attacks; a question naming tyres fired on every 'tire' line; the 4,096-token window blocks a 10k-token policy.\n- **Vela 2.0 (vLLM Semantic Router project (with KR Labs))**: noul plus trained safety, PII and hallucination span heads with AUC over 14 safety sets, but issue 6298 found the trained adapters missed attacks the base model refused.\n\n- not this job: Laya (Convai Innovations Pvt. Ltd.): noul exists, but Arize scored Laya 57.3 percent on evaluator suites with calibration error 0.330 and a toxic-content question flipped from 0.11 to 0.87 AUC on rewording; an HN tester saw a die roll of 3 come back 9 percent odd at 91 percent confidence.\n- not this job: GLiClass (Knowledgator Engineering Ltd.): No predicate type with a calibrated P(yes); labels get softmax or independent sigmoid scores, F1 is the only published metric and issue 12 shows saturation at 0.99.\n\n1 listings in the category were added after this reading and aren't marked yet.\n\n## What stops agents finishing\n\n- disguised or encoded attacks pass a text-only gate\n- threshold at 0.9 blocks every legitimate action, at 0.5 lets unauthorised ones through\n- 4,096-token window too short for a long fixed policy\n\n## Around it\n\nBefore: agent-frameworks.propose-tool-call, chat.receive-message. After: sandboxes.run-command, human-review, agent-frameworks.refuse-tool-call.\n\n## Where this comes from\n\nWeight 16 of 100 in Decision models: 8 of 41 accounts; the AWS nine-desk test, two GitHub issues on wrong yeses, the Vela guard issue and the Dataiku exception threshold work\n\nAs data: https://www.anchorterminal.com/api/v1/use-cases/decision-models.gate-action.json, or the MCP tool get_use_case with id decision-models.gate-action. Draft, read 2026-11-12 from public docs and first-hand accounts, not yet reviewed by a person.\n",
  "meta": {
    "attribution": "Anchor Terminal (https://www.anchorterminal.com)",
    "attributionRequired": true,
    "credit": "CC BY 4.0, credit Anchor Terminal and link the page or API address you took it from (https://www.anchorterminal.com/terms/#data)",
    "docs": "https://www.anchorterminal.com/docs/",
    "generatedAt": "2026-11-16",
    "license": "CC-BY-4.0",
    "method": "https://www.anchorterminal.com/benchmark/",
    "methodology": "0.8",
    "openapi": "https://www.anchorterminal.com/openapi.json",
    "preview": false,
    "run": "2026-10-01",
    "runLabel": "October 2026 research run",
    "terms": "https://www.anchorterminal.com/terms/#data"
  },
  "page": {
    "breadcrumbs": [
      {
        "name": "Home",
        "url": "https://www.anchorterminal.com/"
      },
      {
        "name": "For agents",
        "url": "https://www.anchorterminal.com/agents/"
      },
      {
        "name": "Niches",
        "url": "https://www.anchorterminal.com/niches/"
      },
      {
        "name": "Yes or no gate before an action runs",
        "url": ""
      }
    ],
    "description": "Yes or no gate before an action runs: 9 companies fully do it, 12 with a limit, 42 carry the tag. Every listing in Decision models with the reason and a source.",
    "facts": null,
    "h1": "Yes or no gate before an action runs",
    "image": "https://www.anchorterminal.com/assets/og/default.png",
    "path": "/niches/decision-models.gate-action/",
    "published": "",
    "section": "agents",
    "title": "Yes or no gate before an action runs: who does it | Anchor Terminal",
    "toc": null,
    "updated": "2026-11-15",
    "url": "https://www.anchorterminal.com/niches/decision-models.gate-action/"
  },
  "tokens": {
    "markdown": 2100,
    "slim": 430
  },
  "version": 1
}
