{"id":"2764d98a2015a2aa111c633f705471ae3ca0df5333c6000f861efc1c2008b1f7","title":"Beyond Plain MCP: Do Decision Models and Code Mode Make Tool Use Cheaper and Faster?","argument":"Abhishek Rai’s paper reports a controlled evaluation of ways to help language model agents select and use tools through Model Context Protocol (MCP) gateways. It compares plain MCP with code mode, where agents write scripts against tool stubs, and decision layers, which preselect tool subsets. Across a harness of 12 servers and three open-weight agents, the paper reports that code mode cut token usage by 70 to 90 percent while boosting utility. It also finds that declaring return-shape hints in code mode stubs could swing success by up to 39.5 points for lower-performing agents—an effect the paper says was larger than adding a router. The central hypothesis that a decision layer in front of code mode beats code mode alone was not confirmed: reported improvements for weaker agents were minor and statistically non-significant, while the strongest agent had a slight cost. This makes the paper’s practical point more specific than “add a router”: in its evaluation, how tool outputs are described in code mode stubs mattered substantially. The results are limited to the tested harness and agents; the abstract does not establish how they transfer to other tool catalogs or agents. Rai says the paper releases its harness, a finetuned 4B tool selection model, and evaluation traces to support further experimentation.","topicIds":["intelligence-and-discovery"],"domain":"science","claimType":"attributed reading","limits":"The reported evaluation uses a controlled harness across 12 servers and three open-weight agents; the supplied abstract does not establish whether the findings transfer to other tool catalogs or agents. The supplied evidence is the paper’s abstract, not its full methods or results, so details beyond the abstract cannot be assessed here.","speaker":{"name":"Rai, Abhishek","roleAtSource":"Article author"},"source":{"title":"Beyond Plain MCP: Do Decision Models and Code Mode Make Tool Use Cheaper and Faster?","publisher":"arXiv cs.AI","format":"Article","url":"https://arxiv.org/abs/2611.19123","publishedAt":"2026-11-14","dateBasis":"Date basis not provided."},"evidence":[{"text":"The paper evaluates code mode and decision layers against plain MCP in a controlled harness.","excerpt":"Using a controlled harness across 12 servers, up to 1,958 distractor tools, and three open-weight agents including Qwen, Gemma, and Nemotron","url":"https://arxiv.org/abs/2611.19123","relationship":"Source-stated"},{"text":"The paper reports that code mode reduced token usage and boosted utility compared with plain MCP.","excerpt":"Code mode substantially outperforms plain MCP, cutting token usage by 70 to 90 percent while significantly boosting utility.","url":"https://arxiv.org/abs/2611.19123","relationship":"Source-stated"},{"text":"The paper says return-shape hints in code mode stubs strongly affected success for lower-performing agents.","excerpt":"Declaring return-shape hints in code mode stubs is a first-order variable, swinging success by up to 39.5 points for lower-performing agents","url":"https://arxiv.org/abs/2611.19123","relationship":"Source-stated"},{"text":"The paper did not confirm its hypothesis that putting a decision layer before code mode beats code mode alone.","excerpt":"The central hypothesis that a decision layer in front of code mode beats code mode alone is not confirmed","url":"https://arxiv.org/abs/2611.19123","relationship":"Source-stated"},{"text":"The paper reports minor, statistically non-significant improvements for weaker agents and a slight cost for the strongest agent.","excerpt":"showing minor, statistically non-significant improvements for weaker agents and a slight cost for the strongest.","url":"https://arxiv.org/abs/2611.19123","relationship":"Source-stated"},{"text":"The abstract describes evaluation evidence from a specific harness and three open-weight agents; whether the findings transfer beyond them remains untested in the supplied evidence.","excerpt":"Using a controlled harness across 12 servers, up to 1,958 distractor tools, and three open-weight agents including Qwen, Gemma, and Nemotron","url":"https://arxiv.org/abs/2611.19123","relationship":"Escape Meridian interpretation"}],"sourceMedia":{"status":"not_found","sourceUrl":"https://arxiv.org/abs/2611.19123","checkedAt":"2026-11-17T13:37:00.096Z","items":[],"note":"No suitable media was found by the supported page extraction."},"assumptions":[],"openQuestion":{"attribution":"Open question","text":"Do the reported advantages of code mode, and the findings about return-shape hints and decision layers, hold with other tool catalogs and agents?"},"whatWouldChangeView":{"attribution":"Change condition not recorded","text":""},"editorialComparisons":[],"live":{"title":"Beyond Plain MCP: Do Decision Models and Code Mode Make Tool Use Cheaper and Faster?","summary":"Abhishek Rai’s paper reports a controlled evaluation of ways to help language model agents select and use tools through Model Context Protocol (MCP) gateways. It compares plain MCP with code mode, where agents write scripts against tool stubs, and decision layers, which preselect tool subsets. Across a harness of 12 servers and three open-weight agents, the paper reports that code mode cut token usage by 70 to 90 percent while boosting utility. It also finds that declaring return-shape hints in code mode stubs could swing success by up to 39.5 points for lower-performing agents—an effect the paper says was larger than adding a router. The central hypothesis that a decision layer in front of code mode beats code mode alone was not confirmed: reported improvements for weaker agents were minor and statistically non-significant, while the strongest agent had a slight cost. This makes the paper’s practical point more specific than “add a router”: in its evaluation, how tool outputs are described in code mode stubs mattered substantially. The results are limited to the tested harness and agents; the abstract does not establish how they transfer to other tool catalogs or agents. Rai says the paper releases its harness, a finetuned 4B tool selection model, and evaluation traces to support further experimentation.","question":"Do the reported advantages of code mode, and the findings about return-shape hints and decision layers, hold with other tool catalogs and agents?","claims":[{"text":"The paper evaluates code mode and decision layers against plain MCP in a controlled harness.","excerpt":"Using a controlled harness across 12 servers, up to 1,958 distractor tools, and three open-weight agents including Qwen, Gemma, and Nemotron","support":"source-stated"},{"text":"The paper reports that code mode reduced token usage and boosted utility compared with plain MCP.","excerpt":"Code mode substantially outperforms plain MCP, cutting token usage by 70 to 90 percent while significantly boosting utility.","support":"source-stated"},{"text":"The paper says return-shape hints in code mode stubs strongly affected success for lower-performing agents.","excerpt":"Declaring return-shape hints in code mode stubs is a first-order variable, swinging success by up to 39.5 points for lower-performing agents","support":"source-stated"},{"text":"The paper did not confirm its hypothesis that putting a decision layer before code mode beats code mode alone.","excerpt":"The central hypothesis that a decision layer in front of code mode beats code mode alone is not confirmed","support":"source-stated"},{"text":"The paper reports minor, statistically non-significant improvements for weaker agents and a slight cost for the strongest agent.","excerpt":"showing minor, statistically non-significant improvements for weaker agents and a slight cost for the strongest.","support":"source-stated"},{"text":"The abstract describes evaluation evidence from a specific harness and three open-weight agents; whether the findings transfer beyond them remains untested in the supplied evidence.","excerpt":"Using a controlled harness across 12 servers, up to 1,958 distractor tools, and three open-weight agents including Qwen, Gemma, and Nemotron","support":"interpretation"}],"limitations":["The reported evaluation uses a controlled harness across 12 servers and three open-weight agents; the supplied abstract does not establish whether the findings transfer to other tool catalogs or agents.","The supplied evidence is the paper’s abstract, not its full methods or results, so details beyond the abstract cannot be assessed here."],"mediaId":"6fbc8b740953e302b4f03924ed498d3d14214f417e7ebca4a849eb185bf0fbb1","correctionReason":null,"id":"2764d98a2015a2aa111c633f705471ae3ca0df5333c6000f861efc1c2008b1f7","publicationId":"375e47d9-c155-428e-84db-840c091cc828","version":1,"sourceUrl":"https://arxiv.org/abs/2611.19123","sourceTitle":"Beyond Plain MCP: Do Decision Models and Code Mode Make Tool Use Cheaper and Faster?","publisher":"arXiv cs.AI","publishedAt":null,"originalPublished":{"value":"2026-11-14","precision":"day","basis":"publisher_citation_date"},"firstObservedAt":"2026-11-17T08:27:59.415Z","createdAt":"2026-11-17T13:05:38.316Z","historical":true,"entityIds":["arxiv"],"topicIds":["intelligence-and-discovery"],"media":[{"id":"6fbc8b740953e302b4f03924ed498d3d14214f417e7ebca4a849eb185bf0fbb1","kind":"article","url":"https://arxiv.org/abs/2611.19123","sourceUrl":"https://arxiv.org/abs/2611.19123","label":"Read original","provenance":"source_link","verification":{"status":"verified","checkedAt":"2026-11-17T13:04:30.200Z","method":"acquired_source","contentType":"text/html"},"recommendationReason":"Recommended by Jev for understanding this source; checked by Luna."}],"sourceMedia":{"status":"not_found","sourceUrl":"https://arxiv.org/abs/2611.19123","checkedAt":"2026-11-17T13:37:00.096Z","items":[],"note":"No suitable media was found by the supported page extraction."},"editorialDecision":{"outcome":"accept","reasonCodes":["material_development"],"rubricVersion":"stream-admission-v9","mediaId":"6fbc8b740953e302b4f03924ed498d3d14214f417e7ebca4a849eb185bf0fbb1","confidence":0.5800000000000001,"provider":"typesafe","model":"jev-1.13.0","inputHash":"43811369fd6345852377a91bd96ee535a3480bebaa6577f3a8237b4efb24b04d","decidedAt":"2026-11-17T13:04:33.824Z"},"provenance":{"textHash":"f27a59cdd44041df714aad7757c6893f5638872df1786f33a6905adb039253cd","observationId":"16ff89d8cd1d0c709f58cfe6170494ec5e26adf7b8c30bd0e4d1d3c63dd84e90","retrievedAt":"2026-11-17T13:04:29.689Z","kind":"publisher-page","preparer":{"model":"gpt-6-luna","runId":"4e5b286f-7a19-4f02-81a4-b6e71c552b66"},"reviewer":{"model":"gpt-6-luna","runId":"6742aeb7-8db2-4d18-8bf0-85d08eed074d"}},"audit":{"status":"passed","reviewedAt":"2026-11-17T13:05:38.316Z","model":"anthropic/claude-opus-5.5"},"sourceAttribution":{"authors":[{"name":"Rai, Abhishek","url":null,"kind":"unknown"}],"authorRole":"article-author","hosts":[],"status":"stated","authorRoster":{"truncated":false,"omitted":0,"assertionGroupsOmitted":0},"participation":[{"actor":{"name":"Rai, Abhishek","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Rai, Abhishek","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.19123","locator":"head/meta[name=author|citation_author]"},"assertionId":"41052d6b1d5be8fb81e1a10e7bd7c0a84e97db5c1683ac520c5915b6380ce0cc:participation:0","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-17T13:37:00.096Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"41052d6b1d5be8fb81e1a10e7bd7c0a84e97db5c1683ac520c5915b6380ce0cc","sourceUrl":"https://arxiv.org/abs/2611.19123","contentSha256":"8c3af4730d5031b155f6a2ce1c6516b8785a6e4269bccc7f80247eabf5af3ea7","textHash":"f27a59cdd44041df714aad7757c6893f5638872df1786f33a6905adb039253cd","observedAt":"2026-11-17T13:37:00.096Z"}}],"participationCoverage":{"total":1,"returned":1,"complete":true,"fullAssertions":"source-metadata-endpoint"},"distribution":null,"publisher":null,"sourceUrl":"https://arxiv.org/abs/2611.19123","snapshotId":"41052d6b1d5be8fb81e1a10e7bd7c0a84e97db5c1683ac520c5915b6380ce0cc","contentSha256":"8c3af4730d5031b155f6a2ce1c6516b8785a6e4269bccc7f80247eabf5af3ea7","textHash":"f27a59cdd44041df714aad7757c6893f5638872df1786f33a6905adb039253cd","observedAt":"2026-11-17T13:37:00.096Z","basis":"publisher-page-metadata","sourceLinks":41,"linksTruncated":false,"metadataUrl":"/api/pipeline/stream/2764d98a2015a2aa111c633f705471ae3ca0df5333c6000f861efc1c2008b1f7/source"}},"canonicalUrl":"https://escapemeridian.com/readings/beyond-plain-mcp-do-decision-models-and-code-mode-make-tool-use-2764d98a","formats":{"html":"/readings/beyond-plain-mcp-do-decision-models-and-code-mode-make-tool-use-2764d98a","markdown":"/readings/beyond-plain-mcp-do-decision-models-and-code-mode-make-tool-use-2764d98a.md","json":"/readings/beyond-plain-mcp-do-decision-models-and-code-mode-make-tool-use-2764d98a.json"},"attribution":"Headline and summary by Escape Meridian, prepared by Luna from the attributed source. A source check is not independent confirmation of the source’s claims.","maintainedGuideLinks":[],"conditions":[],"readerContext":{"version":"reader-context-v1","attribution":"Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement.","terms":[{"id":"term-tool-use","name":"Tool use","explanation":"A model's ability to operate other software, such as a search engine, a calculator or a code runner, while it works.","referenceUrl":"https://docs.anthropic.com/en/docs/build-with-claude/tool-use/overview"},{"id":"term-open-weights","name":"Open weights","explanation":"A model whose trained numbers are published so others can download, run and inspect it; the license and training data may still be restricted.","referenceUrl":"https://opensource.org/ai/open-weights"},{"id":"term-fine-tuning","name":"Fine-tuning","explanation":"Further training of an existing model to adapt it to a particular task or use.","referenceUrl":"https://developers.google.com/machine-learning/glossary#fine-tuning"}],"entities":[{"id":"arxiv","name":"arXiv","kind":"organization","context":"arXiv is a research manuscript repository where authors share papers that may not yet have been peer reviewed.","referenceUrls":["https://arxiv.org/","https://info.arxiv.org/help/api/index.html"],"path":"/entities/arxiv","json":"/entities/arxiv.json","markdown":"/entities/arxiv.md"}]},"researchQuestions":[],"withdrawn":false,"references":{"schemaVersion":1,"target":{"type":"reading","id":"2764d98a2015a2aa111c633f705471ae3ca0df5333c6000f861efc1c2008b1f7","revision":"375e47d9-c155-428e-84db-840c091cc828","revisionKind":"publication-id","title":"Beyond Plain MCP: Do Decision Models and Code Mode Make Tool Use Cheaper and Faster?","url":"https://escapemeridian.com/readings/beyond-plain-mcp-do-decision-models-and-code-mode-make-tool-use-2764d98a"},"source":{"type":"source","id":"https://arxiv.org/abs/2611.19123","revision":"f27a59cdd44041df714aad7757c6893f5638872df1786f33a6905adb039253cd","revisionKind":"sha256-source-text","title":"Beyond Plain MCP: Do Decision Models and Code Mode Make Tool Use Cheaper and Faster?","url":"https://arxiv.org/abs/2611.19123"},"status":"available","items":[],"next":null,"scope":"Explicit citations in public readings to this reading. Shared topics are not citations. Other record types need an explicit reference adapter."},"revisions":[{"id":"2764d98a2015a2aa111c633f705471ae3ca0df5333c6000f861efc1c2008b1f7","title":"Beyond Plain MCP: Do Decision Models and Code Mode Make Tool Use Cheaper and Faster?","argument":"Abhishek Rai’s paper reports a controlled evaluation of ways to help language model agents select and use tools through Model Context Protocol (MCP) gateways. It compares plain MCP with code mode, where agents write scripts against tool stubs, and decision layers, which preselect tool subsets. Across a harness of 12 servers and three open-weight agents, the paper reports that code mode cut token usage by 70 to 90 percent while boosting utility. It also finds that declaring return-shape hints in code mode stubs could swing success by up to 39.5 points for lower-performing agents—an effect the paper says was larger than adding a router. The central hypothesis that a decision layer in front of code mode beats code mode alone was not confirmed: reported improvements for weaker agents were minor and statistically non-significant, while the strongest agent had a slight cost. This makes the paper’s practical point more specific than “add a router”: in its evaluation, how tool outputs are described in code mode stubs mattered substantially. The results are limited to the tested harness and agents; the abstract does not establish how they transfer to other tool catalogs or agents. Rai says the paper releases its harness, a finetuned 4B tool selection model, and evaluation traces to support further experimentation.","topicIds":["intelligence-and-discovery"],"domain":"science","claimType":"attributed reading","limits":"The reported evaluation uses a controlled harness across 12 servers and three open-weight agents; the supplied abstract does not establish whether the findings transfer to other tool catalogs or agents. The supplied evidence is the paper’s abstract, not its full methods or results, so details beyond the abstract cannot be assessed here.","speaker":{"name":"Rai, Abhishek","roleAtSource":"Article author"},"source":{"title":"Beyond Plain MCP: Do Decision Models and Code Mode Make Tool Use Cheaper and Faster?","publisher":"arXiv cs.AI","format":"Article","url":"https://arxiv.org/abs/2611.19123","publishedAt":"2026-11-14","dateBasis":"Date basis not provided."},"evidence":[{"text":"The paper evaluates code mode and decision layers against plain MCP in a controlled harness.","excerpt":"Using a controlled harness across 12 servers, up to 1,958 distractor tools, and three open-weight agents including Qwen, Gemma, and Nemotron","url":"https://arxiv.org/abs/2611.19123","relationship":"Source-stated"},{"text":"The paper reports that code mode reduced token usage and boosted utility compared with plain MCP.","excerpt":"Code mode substantially outperforms plain MCP, cutting token usage by 70 to 90 percent while significantly boosting utility.","url":"https://arxiv.org/abs/2611.19123","relationship":"Source-stated"},{"text":"The paper says return-shape hints in code mode stubs strongly affected success for lower-performing agents.","excerpt":"Declaring return-shape hints in code mode stubs is a first-order variable, swinging success by up to 39.5 points for lower-performing agents","url":"https://arxiv.org/abs/2611.19123","relationship":"Source-stated"},{"text":"The paper did not confirm its hypothesis that putting a decision layer before code mode beats code mode alone.","excerpt":"The central hypothesis that a decision layer in front of code mode beats code mode alone is not confirmed","url":"https://arxiv.org/abs/2611.19123","relationship":"Source-stated"},{"text":"The paper reports minor, statistically non-significant improvements for weaker agents and a slight cost for the strongest agent.","excerpt":"showing minor, statistically non-significant improvements for weaker agents and a slight cost for the strongest.","url":"https://arxiv.org/abs/2611.19123","relationship":"Source-stated"},{"text":"The abstract describes evaluation evidence from a specific harness and three open-weight agents; whether the findings transfer beyond them remains untested in the supplied evidence.","excerpt":"Using a controlled harness across 12 servers, up to 1,958 distractor tools, and three open-weight agents including Qwen, Gemma, and Nemotron","url":"https://arxiv.org/abs/2611.19123","relationship":"Escape Meridian interpretation"}],"sourceMedia":{"status":"not_found","sourceUrl":"https://arxiv.org/abs/2611.19123","checkedAt":"2026-11-17T13:37:00.096Z","items":[],"note":"No suitable media was found by the supported page extraction."},"assumptions":[],"openQuestion":{"attribution":"Open question","text":"Do the reported advantages of code mode, and the findings about return-shape hints and decision layers, hold with other tool catalogs and agents?"},"whatWouldChangeView":{"attribution":"Change condition not recorded","text":""},"editorialComparisons":[],"live":{"title":"Beyond Plain MCP: Do Decision Models and Code Mode Make Tool Use Cheaper and Faster?","summary":"Abhishek Rai’s paper reports a controlled evaluation of ways to help language model agents select and use tools through Model Context Protocol (MCP) gateways. It compares plain MCP with code mode, where agents write scripts against tool stubs, and decision layers, which preselect tool subsets. Across a harness of 12 servers and three open-weight agents, the paper reports that code mode cut token usage by 70 to 90 percent while boosting utility. It also finds that declaring return-shape hints in code mode stubs could swing success by up to 39.5 points for lower-performing agents—an effect the paper says was larger than adding a router. The central hypothesis that a decision layer in front of code mode beats code mode alone was not confirmed: reported improvements for weaker agents were minor and statistically non-significant, while the strongest agent had a slight cost. This makes the paper’s practical point more specific than “add a router”: in its evaluation, how tool outputs are described in code mode stubs mattered substantially. The results are limited to the tested harness and agents; the abstract does not establish how they transfer to other tool catalogs or agents. Rai says the paper releases its harness, a finetuned 4B tool selection model, and evaluation traces to support further experimentation.","question":"Do the reported advantages of code mode, and the findings about return-shape hints and decision layers, hold with other tool catalogs and agents?","claims":[{"text":"The paper evaluates code mode and decision layers against plain MCP in a controlled harness.","excerpt":"Using a controlled harness across 12 servers, up to 1,958 distractor tools, and three open-weight agents including Qwen, Gemma, and Nemotron","support":"source-stated"},{"text":"The paper reports that code mode reduced token usage and boosted utility compared with plain MCP.","excerpt":"Code mode substantially outperforms plain MCP, cutting token usage by 70 to 90 percent while significantly boosting utility.","support":"source-stated"},{"text":"The paper says return-shape hints in code mode stubs strongly affected success for lower-performing agents.","excerpt":"Declaring return-shape hints in code mode stubs is a first-order variable, swinging success by up to 39.5 points for lower-performing agents","support":"source-stated"},{"text":"The paper did not confirm its hypothesis that putting a decision layer before code mode beats code mode alone.","excerpt":"The central hypothesis that a decision layer in front of code mode beats code mode alone is not confirmed","support":"source-stated"},{"text":"The paper reports minor, statistically non-significant improvements for weaker agents and a slight cost for the strongest agent.","excerpt":"showing minor, statistically non-significant improvements for weaker agents and a slight cost for the strongest.","support":"source-stated"},{"text":"The abstract describes evaluation evidence from a specific harness and three open-weight agents; whether the findings transfer beyond them remains untested in the supplied evidence.","excerpt":"Using a controlled harness across 12 servers, up to 1,958 distractor tools, and three open-weight agents including Qwen, Gemma, and Nemotron","support":"interpretation"}],"limitations":["The reported evaluation uses a controlled harness across 12 servers and three open-weight agents; the supplied abstract does not establish whether the findings transfer to other tool catalogs or agents.","The supplied evidence is the paper’s abstract, not its full methods or results, so details beyond the abstract cannot be assessed here."],"mediaId":"6fbc8b740953e302b4f03924ed498d3d14214f417e7ebca4a849eb185bf0fbb1","correctionReason":null,"id":"2764d98a2015a2aa111c633f705471ae3ca0df5333c6000f861efc1c2008b1f7","publicationId":"375e47d9-c155-428e-84db-840c091cc828","version":1,"sourceUrl":"https://arxiv.org/abs/2611.19123","sourceTitle":"Beyond Plain MCP: Do Decision Models and Code Mode Make Tool Use Cheaper and Faster?","publisher":"arXiv cs.AI","publishedAt":null,"originalPublished":{"value":"2026-11-14","precision":"day","basis":"publisher_citation_date"},"firstObservedAt":"2026-11-17T08:27:59.415Z","createdAt":"2026-11-17T13:05:38.316Z","historical":true,"entityIds":["arxiv"],"topicIds":["intelligence-and-discovery"],"media":[{"id":"6fbc8b740953e302b4f03924ed498d3d14214f417e7ebca4a849eb185bf0fbb1","kind":"article","url":"https://arxiv.org/abs/2611.19123","sourceUrl":"https://arxiv.org/abs/2611.19123","label":"Read original","provenance":"source_link","verification":{"status":"verified","checkedAt":"2026-11-17T13:04:30.200Z","method":"acquired_source","contentType":"text/html"},"recommendationReason":"Recommended by Jev for understanding this source; checked by Luna."}],"sourceMedia":{"status":"not_found","sourceUrl":"https://arxiv.org/abs/2611.19123","checkedAt":"2026-11-17T13:37:00.096Z","items":[],"note":"No suitable media was found by the supported page extraction."},"editorialDecision":{"outcome":"accept","reasonCodes":["material_development"],"rubricVersion":"stream-admission-v9","mediaId":"6fbc8b740953e302b4f03924ed498d3d14214f417e7ebca4a849eb185bf0fbb1","confidence":0.5800000000000001,"provider":"typesafe","model":"jev-1.13.0","inputHash":"43811369fd6345852377a91bd96ee535a3480bebaa6577f3a8237b4efb24b04d","decidedAt":"2026-11-17T13:04:33.824Z"},"provenance":{"textHash":"f27a59cdd44041df714aad7757c6893f5638872df1786f33a6905adb039253cd","observationId":"16ff89d8cd1d0c709f58cfe6170494ec5e26adf7b8c30bd0e4d1d3c63dd84e90","retrievedAt":"2026-11-17T13:04:29.689Z","kind":"publisher-page","preparer":{"model":"gpt-6-luna","runId":"4e5b286f-7a19-4f02-81a4-b6e71c552b66"},"reviewer":{"model":"gpt-6-luna","runId":"6742aeb7-8db2-4d18-8bf0-85d08eed074d"}},"audit":{"status":"passed","reviewedAt":"2026-11-17T13:05:38.316Z","model":"anthropic/claude-opus-5.5"},"sourceAttribution":{"authors":[{"name":"Rai, Abhishek","url":null,"kind":"unknown"}],"authorRole":"article-author","hosts":[],"status":"stated","authorRoster":{"truncated":false,"omitted":0,"assertionGroupsOmitted":0},"participation":[{"actor":{"name":"Rai, Abhishek","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Rai, Abhishek","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.19123","locator":"head/meta[name=author|citation_author]"},"assertionId":"41052d6b1d5be8fb81e1a10e7bd7c0a84e97db5c1683ac520c5915b6380ce0cc:participation:0","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-17T13:37:00.096Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"41052d6b1d5be8fb81e1a10e7bd7c0a84e97db5c1683ac520c5915b6380ce0cc","sourceUrl":"https://arxiv.org/abs/2611.19123","contentSha256":"8c3af4730d5031b155f6a2ce1c6516b8785a6e4269bccc7f80247eabf5af3ea7","textHash":"f27a59cdd44041df714aad7757c6893f5638872df1786f33a6905adb039253cd","observedAt":"2026-11-17T13:37:00.096Z"}}],"participationCoverage":{"total":1,"returned":1,"complete":true,"fullAssertions":"source-metadata-endpoint"},"distribution":null,"publisher":null,"sourceUrl":"https://arxiv.org/abs/2611.19123","snapshotId":"41052d6b1d5be8fb81e1a10e7bd7c0a84e97db5c1683ac520c5915b6380ce0cc","contentSha256":"8c3af4730d5031b155f6a2ce1c6516b8785a6e4269bccc7f80247eabf5af3ea7","textHash":"f27a59cdd44041df714aad7757c6893f5638872df1786f33a6905adb039253cd","observedAt":"2026-11-17T13:37:00.096Z","basis":"publisher-page-metadata","sourceLinks":41,"linksTruncated":false,"metadataUrl":"/api/pipeline/stream/2764d98a2015a2aa111c633f705471ae3ca0df5333c6000f861efc1c2008b1f7/source"}}}],"audits":[{"publicationId":"375e47d9-c155-428e-84db-840c091cc828","verdict":"passed","findings":[{"dimension":"attribution","status":"pass","explanation":"The reading credits the paper to Abhishek Rai, the sole listed author, and keeps findings attributed to the paper ('the paper reports', 'Rai says') rather than presenting them as established fact.","excerpt":"Authors:Abhishek Rai"},{"dimension":"summary","status":"pass","explanation":"The title matches the source. The summary accurately conveys the comparison (plain MCP vs code mode vs decision layers), the 70–90 percent token reduction, the up-to-39.5-point return-shape hint effect, the unconfirmed central hypothesis, and the released artifacts. It explains code mode and decision layers in plain language. Small clarity gap: 'router' appears without noting that the source defines decision layers as cheap routers; tooltips for MCP, tool stubs and token usage would also help. The question follows from the scope limits.","excerpt":"code mode where agents write scripts against tool stubs and decision layers which are cheap routers that preselect tool subsets"},{"dimension":"date","status":"pass","explanation":"originalPublished 2026-11-14 matches the arXiv v1 submission date, not the later feed date of 2026-11-17.","excerpt":"[Submitted on 14 Nov 2026]"},{"dimension":"support","status":"pass","explanation":"Each claim's excerpt appears verbatim in the source and supports it. The transfer-limitation claim is labeled as interpretation and stays conditional. The 'practical point' sentence is qualified with 'in its evaluation' and does not overreach. No benefits or costs are invented.","excerpt":"Code mode substantially outperforms plain MCP, cutting token usage by 70 to 90 percent while significantly boosting utility."},{"dimension":"media","status":"pass","explanation":"The selected media is the verified direct arXiv abstract page for this paper, which is the best route to the original source.","excerpt":null}],"createdAt":"2026-11-17T13:05:38.316Z","model":"anthropic/claude-opus-5.5"}],"correctionUrl":"https://escapemeridian.com/corrections/"}