{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:Y2MI66EMP5WWQKKIEDXD2HARDL","short_pith_number":"pith:Y2MI66EM","schema_version":"1.0","canonical_sha256":"c6988f788c7f6d68294820ee3d1c111afa22661657c9e85f4f4bcc9c8e0f7b0a","source":{"kind":"arxiv","id":"2607.20478","version":1},"attestation_state":"computed","paper":{"title":"Verifier-First Evaluation of Agentic LLMs for Infrastructure-as-Code Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.SE","authors_text":"Mohamed Jouini","submitted_at":"2026-05-29T12:43:27Z","abstract_excerpt":"Infrastructure-as-Code (IaC) generation from natural language requires satisfying provider schemas, dependency planning, and organizational policy constraints, not merely producing syntactically plausible configurations. We present a verifier-first empirical study of seven agentic strategies for Terraform generation evaluated on IaC-Eval v2, a modernized 186-task AWS/Terraform benchmark with Rego v1 intent policies. Our evaluation separates failures into three verifier stages (terraform validate, terraform plan, opa eval) and applies McNemar's test with Wilson confidence intervals on all pairw"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2607.20478","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SE","submitted_at":"2026-05-29T12:43:27Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"9ec79cd7daad41d72bc5b024a555955cb4b07c3f850e0d600652f442dafd3bad","abstract_canon_sha256":"879dcf358f4e1594b8cf225dc5d9005a154f6554c00475951e15d20b7ebf023a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-24T00:23:17.812007Z","signature_b64":"kJuuSlFE4rjUgri8eRwCWHPbFI/NA1tF93yJF/8jXtlMaXYGy2ic4NmLPoSIb6wuP7aqV51+lTT2CgCwDD6ADA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c6988f788c7f6d68294820ee3d1c111afa22661657c9e85f4f4bcc9c8e0f7b0a","last_reissued_at":"2026-07-24T00:23:17.811112Z","signature_status":"signed_v1","first_computed_at":"2026-07-24T00:23:17.811112Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Verifier-First Evaluation of Agentic LLMs for Infrastructure-as-Code Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.SE","authors_text":"Mohamed Jouini","submitted_at":"2026-05-29T12:43:27Z","abstract_excerpt":"Infrastructure-as-Code (IaC) generation from natural language requires satisfying provider schemas, dependency planning, and organizational policy constraints, not merely producing syntactically plausible configurations. We present a verifier-first empirical study of seven agentic strategies for Terraform generation evaluated on IaC-Eval v2, a modernized 186-task AWS/Terraform benchmark with Rego v1 intent policies. Our evaluation separates failures into three verifier stages (terraform validate, terraform plan, opa eval) and applies McNemar's test with Wilson confidence intervals on all pairw"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.20478","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2607.20478/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2607.20478","created_at":"2026-07-24T00:23:17.811588+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.20478v1","created_at":"2026-07-24T00:23:17.811588+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.20478","created_at":"2026-07-24T00:23:17.811588+00:00"},{"alias_kind":"pith_short_12","alias_value":"Y2MI66EMP5WW","created_at":"2026-07-24T00:23:17.811588+00:00"},{"alias_kind":"pith_short_16","alias_value":"Y2MI66EMP5WWQKKI","created_at":"2026-07-24T00:23:17.811588+00:00"},{"alias_kind":"pith_short_8","alias_value":"Y2MI66EM","created_at":"2026-07-24T00:23:17.811588+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/Y2MI66EMP5WWQKKIEDXD2HARDL","json":"https://pith.science/pith/Y2MI66EMP5WWQKKIEDXD2HARDL.json","graph_json":"https://pith.science/api/pith-number/Y2MI66EMP5WWQKKIEDXD2HARDL/graph.json","events_json":"https://pith.science/api/pith-number/Y2MI66EMP5WWQKKIEDXD2HARDL/events.json","paper":"https://pith.science/paper/Y2MI66EM"},"agent_actions":{"view_html":"https://pith.science/pith/Y2MI66EMP5WWQKKIEDXD2HARDL","download_json":"https://pith.science/pith/Y2MI66EMP5WWQKKIEDXD2HARDL.json","view_paper":"https://pith.science/paper/Y2MI66EM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.20478&json=true","fetch_graph":"https://pith.science/api/pith-number/Y2MI66EMP5WWQKKIEDXD2HARDL/graph.json","fetch_events":"https://pith.science/api/pith-number/Y2MI66EMP5WWQKKIEDXD2HARDL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Y2MI66EMP5WWQKKIEDXD2HARDL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Y2MI66EMP5WWQKKIEDXD2HARDL/action/storage_attestation","attest_author":"https://pith.science/pith/Y2MI66EMP5WWQKKIEDXD2HARDL/action/author_attestation","sign_citation":"https://pith.science/pith/Y2MI66EMP5WWQKKIEDXD2HARDL/action/citation_signature","submit_replication":"https://pith.science/pith/Y2MI66EMP5WWQKKIEDXD2HARDL/action/replication_record"}},"created_at":"2026-07-24T00:23:17.811588+00:00","updated_at":"2026-07-24T00:23:17.811588+00:00"}