{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:M6CVW4LBC2CVX362UAZLAB63U2","short_pith_number":"pith:M6CVW4LB","schema_version":"1.0","canonical_sha256":"67855b716116855befdaa032b007dba6bbc233bee95230b1f356ccec33f16f8d","source":{"kind":"arxiv","id":"2402.01586","version":4},"attestation_state":"computed","paper":{"title":"TrustAgent: Towards Safe and Trustworthy LLM-based Agents","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG","cs.MA"],"primary_cat":"cs.CL","authors_text":"Mingyu Jin, Ruixiang Tang, Wei Cheng, Wenyue Hua, Xianjun Yang, Yongfeng Zhang, Zelong Li","submitted_at":"2024-02-02T17:26:23Z","abstract_excerpt":"The rise of LLM-based agents shows great potential to revolutionize task planning, capturing significant attention. Given that these agents will be integrated into high-stake domains, ensuring their reliability and safety is crucial. This paper presents an Agent-Constitution-based agent framework, TrustAgent, with a particular focus on improving the LLM-based agent safety. The proposed framework ensures strict adherence to the Agent Constitution through three strategic components: pre-planning strategy which injects safety knowledge to the model before plan generation, in-planning strategy whi"},"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":"2402.01586","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-02-02T17:26:23Z","cross_cats_sorted":["cs.AI","cs.LG","cs.MA"],"title_canon_sha256":"0049e80f22f4c077b33654f5b7aa3ebde9568f283970ff083c9a8df241efb479","abstract_canon_sha256":"1d9f0bb6bef769a6d4071649ffd4529b73bb5dee2b2a3162856ffb0b7503243c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:15:29.526844Z","signature_b64":"iiLMSYq1etcoWVE/gxYj5+9Rm+bdrML2uBnrl6UIGorpZgDPeR7GiCV6fb5chrypkzFAUYOJ+/EVqqVHQXCABA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"67855b716116855befdaa032b007dba6bbc233bee95230b1f356ccec33f16f8d","last_reissued_at":"2026-07-05T09:15:29.526327Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:15:29.526327Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"TrustAgent: Towards Safe and Trustworthy LLM-based Agents","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG","cs.MA"],"primary_cat":"cs.CL","authors_text":"Mingyu Jin, Ruixiang Tang, Wei Cheng, Wenyue Hua, Xianjun Yang, Yongfeng Zhang, Zelong Li","submitted_at":"2024-02-02T17:26:23Z","abstract_excerpt":"The rise of LLM-based agents shows great potential to revolutionize task planning, capturing significant attention. Given that these agents will be integrated into high-stake domains, ensuring their reliability and safety is crucial. This paper presents an Agent-Constitution-based agent framework, TrustAgent, with a particular focus on improving the LLM-based agent safety. The proposed framework ensures strict adherence to the Agent Constitution through three strategic components: pre-planning strategy which injects safety knowledge to the model before plan generation, in-planning strategy whi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.01586","kind":"arxiv","version":4},"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/2402.01586/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":"2402.01586","created_at":"2026-07-05T09:15:29.526392+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.01586v4","created_at":"2026-07-05T09:15:29.526392+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.01586","created_at":"2026-07-05T09:15:29.526392+00:00"},{"alias_kind":"pith_short_12","alias_value":"M6CVW4LBC2CV","created_at":"2026-07-05T09:15:29.526392+00:00"},{"alias_kind":"pith_short_16","alias_value":"M6CVW4LBC2CVX362","created_at":"2026-07-05T09:15:29.526392+00:00"},{"alias_kind":"pith_short_8","alias_value":"M6CVW4LB","created_at":"2026-07-05T09:15:29.526392+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.23075","citing_title":"Safety in Self-Evolving LLM Agent Systems: Threats, Amplification, and Case Studies","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2605.16282","citing_title":"Taxonomy and Consistency Analysis of Safety Benchmarks for AI Agents","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2605.04785","citing_title":"AgentTrust: Runtime Safety Evaluation and Interception for AI Agent Tool Use","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2604.18658","citing_title":"Owner-Harm: A Missing Threat Model for AI Agent Safety","ref_index":8,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/M6CVW4LBC2CVX362UAZLAB63U2","json":"https://pith.science/pith/M6CVW4LBC2CVX362UAZLAB63U2.json","graph_json":"https://pith.science/api/pith-number/M6CVW4LBC2CVX362UAZLAB63U2/graph.json","events_json":"https://pith.science/api/pith-number/M6CVW4LBC2CVX362UAZLAB63U2/events.json","paper":"https://pith.science/paper/M6CVW4LB"},"agent_actions":{"view_html":"https://pith.science/pith/M6CVW4LBC2CVX362UAZLAB63U2","download_json":"https://pith.science/pith/M6CVW4LBC2CVX362UAZLAB63U2.json","view_paper":"https://pith.science/paper/M6CVW4LB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.01586&json=true","fetch_graph":"https://pith.science/api/pith-number/M6CVW4LBC2CVX362UAZLAB63U2/graph.json","fetch_events":"https://pith.science/api/pith-number/M6CVW4LBC2CVX362UAZLAB63U2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/M6CVW4LBC2CVX362UAZLAB63U2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/M6CVW4LBC2CVX362UAZLAB63U2/action/storage_attestation","attest_author":"https://pith.science/pith/M6CVW4LBC2CVX362UAZLAB63U2/action/author_attestation","sign_citation":"https://pith.science/pith/M6CVW4LBC2CVX362UAZLAB63U2/action/citation_signature","submit_replication":"https://pith.science/pith/M6CVW4LBC2CVX362UAZLAB63U2/action/replication_record"}},"created_at":"2026-07-05T09:15:29.526392+00:00","updated_at":"2026-07-05T09:15:29.526392+00:00"}