{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:ZXSPO5EECTYFQO3AZDT7U36ZK6","short_pith_number":"pith:ZXSPO5EE","schema_version":"1.0","canonical_sha256":"cde4f7748414f0583b60c8e7fa6fd957968024f2bf2392f82e5b441865990173","source":{"kind":"arxiv","id":"2402.11208","version":2},"attestation_state":"computed","paper":{"title":"Watch Out for Your Agents! Investigating Backdoor Threats to LLM-Based Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.CR","authors_text":"Jie Zhou, Sishuo Chen, Wenkai Yang, Xiaohan Bi, Xu Sun, Yankai Lin","submitted_at":"2024-02-17T06:48:45Z","abstract_excerpt":"Driven by the rapid development of Large Language Models (LLMs), LLM-based agents have been developed to handle various real-world applications, including finance, healthcare, and shopping, etc. It is crucial to ensure the reliability and security of LLM-based agents during applications. However, the safety issues of LLM-based agents are currently under-explored. In this work, we take the first step to investigate one of the typical safety threats, backdoor attack, to LLM-based agents. We first formulate a general framework of agent backdoor attacks, then we present a thorough analysis of diff"},"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":"2402.11208","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2024-02-17T06:48:45Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"e4419a68754466bb673fee088e7dba9ec40d789d3560d8ba87594aeeac73ade4","abstract_canon_sha256":"9bf1c29edc761705eda1ee42ba7dcb029243e58b8fd78efa0f23a215b911a7be"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:27:52.515029Z","signature_b64":"w5NlbrkomlOrgBCmWMjwiP9veLe85MLKHsDnpQP9/8Ziwjwxktgmlh9hVgX15DE1thV48MnYETtGQsFZzNDFCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cde4f7748414f0583b60c8e7fa6fd957968024f2bf2392f82e5b441865990173","last_reissued_at":"2026-07-05T09:27:52.514568Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:27:52.514568Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Watch Out for Your Agents! Investigating Backdoor Threats to LLM-Based Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.CR","authors_text":"Jie Zhou, Sishuo Chen, Wenkai Yang, Xiaohan Bi, Xu Sun, Yankai Lin","submitted_at":"2024-02-17T06:48:45Z","abstract_excerpt":"Driven by the rapid development of Large Language Models (LLMs), LLM-based agents have been developed to handle various real-world applications, including finance, healthcare, and shopping, etc. It is crucial to ensure the reliability and security of LLM-based agents during applications. However, the safety issues of LLM-based agents are currently under-explored. In this work, we take the first step to investigate one of the typical safety threats, backdoor attack, to LLM-based agents. We first formulate a general framework of agent backdoor attacks, then we present a thorough analysis of diff"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.11208","kind":"arxiv","version":2},"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/2402.11208/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":"2402.11208","created_at":"2026-07-05T09:27:52.514617+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.11208v2","created_at":"2026-07-05T09:27:52.514617+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.11208","created_at":"2026-07-05T09:27:52.514617+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZXSPO5EECTYF","created_at":"2026-07-05T09:27:52.514617+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZXSPO5EECTYFQO3A","created_at":"2026-07-05T09:27:52.514617+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZXSPO5EE","created_at":"2026-07-05T09:27:52.514617+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.24322","citing_title":"Securing LLM-Agent Long-Term Memory Against Poisoning: Non-Malleable, Origin-Bound Authority with Machine-Checked Guarantees","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2606.18356","citing_title":"SafeClawBench: Separating Semantic, Audit-Evidence, and Sandbox Harm in Tool-Using LLM Agents","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2410.02644","citing_title":"Agent Security Bench (ASB): Formalizing and Benchmarking Attacks and Defenses in LLM-based Agents","ref_index":154,"is_internal_anchor":false},{"citing_arxiv_id":"2604.15415","citing_title":"HarmfulSkillBench: How Do Harmful Skills Weaponize Your Agents?","ref_index":74,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZXSPO5EECTYFQO3AZDT7U36ZK6","json":"https://pith.science/pith/ZXSPO5EECTYFQO3AZDT7U36ZK6.json","graph_json":"https://pith.science/api/pith-number/ZXSPO5EECTYFQO3AZDT7U36ZK6/graph.json","events_json":"https://pith.science/api/pith-number/ZXSPO5EECTYFQO3AZDT7U36ZK6/events.json","paper":"https://pith.science/paper/ZXSPO5EE"},"agent_actions":{"view_html":"https://pith.science/pith/ZXSPO5EECTYFQO3AZDT7U36ZK6","download_json":"https://pith.science/pith/ZXSPO5EECTYFQO3AZDT7U36ZK6.json","view_paper":"https://pith.science/paper/ZXSPO5EE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.11208&json=true","fetch_graph":"https://pith.science/api/pith-number/ZXSPO5EECTYFQO3AZDT7U36ZK6/graph.json","fetch_events":"https://pith.science/api/pith-number/ZXSPO5EECTYFQO3AZDT7U36ZK6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZXSPO5EECTYFQO3AZDT7U36ZK6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZXSPO5EECTYFQO3AZDT7U36ZK6/action/storage_attestation","attest_author":"https://pith.science/pith/ZXSPO5EECTYFQO3AZDT7U36ZK6/action/author_attestation","sign_citation":"https://pith.science/pith/ZXSPO5EECTYFQO3AZDT7U36ZK6/action/citation_signature","submit_replication":"https://pith.science/pith/ZXSPO5EECTYFQO3AZDT7U36ZK6/action/replication_record"}},"created_at":"2026-07-05T09:27:52.514617+00:00","updated_at":"2026-07-05T09:27:52.514617+00:00"}