{"paper":{"title":"MIRAGE: Online LLM Simulation for Microservice Dependency Testing","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Online LLM simulation lets microservice tests generate dependency responses at runtime, reaching 99 percent status-code and response-shape fidelity where record-replay reaches only 62 and 16 percent.","cross_cats":[],"primary_cat":"cs.SE","authors_text":"Xinran Zhang","submitted_at":"2026-04-06T16:10:23Z","abstract_excerpt":"Existing approaches to microservice dependency simulation--record-replay, pattern-mining, and specification-driven stubs--generate static artifacts before test execution. These artifacts can only reproduce behaviors encoded at generation time; on error-handling and code-reasoning scenarios, which are underrepresented in typical trace corpora, record-replay achieves 0% and 12% fidelity in our evaluation.\n  We propose online LLM simulation, a runtime approach where the LLM answers each dependency request as it arrives, maintaining cross-request state throughout a test scenario. The model reads t"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"In white-box mode, MIRAGE achieves 99% status-code and 99% response-shape fidelity, compared to 62% / 16% for record-replay. Caller integration tests produce the same pass/fail outcomes with MIRAGE as with real dependencies (8/8 scenarios).","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That an LLM prompted with source code and traces will produce behavior that generalizes to unseen error-handling and reasoning scenarios without introducing hallucinations or inconsistencies that would alter test outcomes.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Online LLM simulation of microservice dependencies achieves 99% status-code and response-shape fidelity across 110 scenarios on three systems, far exceeding record-replay baselines.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Online LLM simulation lets microservice tests generate dependency responses at runtime, reaching 99 percent status-code and response-shape fidelity where record-replay reaches only 62 and 16 percent.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"9d4317d5ce23562f5bcc4e6e05751550fbda2a64be4cc9b17e08d356da59713e"},"source":{"id":"2604.04806","kind":"arxiv","version":4},"verdict":{"id":"36b2400d-c1bc-4994-8981-cb9689ce9696","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-10T19:39:22.817421Z","strongest_claim":"In white-box mode, MIRAGE achieves 99% status-code and 99% response-shape fidelity, compared to 62% / 16% for record-replay. Caller integration tests produce the same pass/fail outcomes with MIRAGE as with real dependencies (8/8 scenarios).","one_line_summary":"Online LLM simulation of microservice dependencies achieves 99% status-code and response-shape fidelity across 110 scenarios on three systems, far exceeding record-replay baselines.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That an LLM prompted with source code and traces will produce behavior that generalizes to unseen error-handling and reasoning scenarios without introducing hallucinations or inconsistencies that would alter test outcomes.","pith_extraction_headline":"Online LLM simulation lets microservice tests generate dependency responses at runtime, reaching 99 percent status-code and response-shape fidelity where record-replay reaches only 62 and 16 percent."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.04806/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}