{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:JBZP6UTXLODYEDYXNRTN5LC5IZ","short_pith_number":"pith:JBZP6UTX","schema_version":"1.0","canonical_sha256":"4872ff52775b87820f176c66deac5d465153faf6dc6ea1f61813c3a1738a2df4","source":{"kind":"arxiv","id":"2504.19956","version":2},"attestation_state":"computed","paper":{"title":"Securing Agentic AI: A Comprehensive Threat Model and Mitigation Framework for Generative AI Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CR","authors_text":"Om Narayan, Vineeth Sai Narajala","submitted_at":"2025-04-28T16:29:24Z","abstract_excerpt":"As generative AI (GenAI) agents become more common in enterprise settings, they introduce security challenges that differ significantly from those posed by traditional systems. These agents are not just LLMs; they reason, remember, and act, often with minimal human oversight. This paper introduces a comprehensive threat model tailored specifically for GenAI agents, focusing on how their autonomy, persistent memory access, complex reasoning, and tool integration create novel risks. This research work identifies 9 primary threats and organizes them across five key domains: cognitive architecture"},"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":"2504.19956","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-04-28T16:29:24Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"a777235e5b709a584c60eab521ca5cb87787d7abce5d9c3d7b5a549a456430cd","abstract_canon_sha256":"0a4e06333ba37d1a7c38973941795d55b0cd11aa066d19e22873a51c71d33dc3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:57:57.285417Z","signature_b64":"l0ARDIkBcp4T1Bc91pxrX7HVnhWuqWbHjomCMFTDuDlwubAZsC58OJ9Sto463pTdzCouRkpeU6Y/kBoD7JHwCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4872ff52775b87820f176c66deac5d465153faf6dc6ea1f61813c3a1738a2df4","last_reissued_at":"2026-07-05T10:57:57.284875Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:57:57.284875Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Securing Agentic AI: A Comprehensive Threat Model and Mitigation Framework for Generative AI Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CR","authors_text":"Om Narayan, Vineeth Sai Narajala","submitted_at":"2025-04-28T16:29:24Z","abstract_excerpt":"As generative AI (GenAI) agents become more common in enterprise settings, they introduce security challenges that differ significantly from those posed by traditional systems. These agents are not just LLMs; they reason, remember, and act, often with minimal human oversight. This paper introduces a comprehensive threat model tailored specifically for GenAI agents, focusing on how their autonomy, persistent memory access, complex reasoning, and tool integration create novel risks. This research work identifies 9 primary threats and organizes them across five key domains: cognitive architecture"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.19956","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/2504.19956/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":"2504.19956","created_at":"2026-07-05T10:57:57.284935+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.19956v2","created_at":"2026-07-05T10:57:57.284935+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.19956","created_at":"2026-07-05T10:57:57.284935+00:00"},{"alias_kind":"pith_short_12","alias_value":"JBZP6UTXLODY","created_at":"2026-07-05T10:57:57.284935+00:00"},{"alias_kind":"pith_short_16","alias_value":"JBZP6UTXLODYEDYX","created_at":"2026-07-05T10:57:57.284935+00:00"},{"alias_kind":"pith_short_8","alias_value":"JBZP6UTX","created_at":"2026-07-05T10:57:57.284935+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":11,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.12797","citing_title":"The Containment Gap: How Deployed Agentic AI Frameworks Fail Public-Facing Safety Requirements","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2606.29142","citing_title":"Agent Security Meets Regulatory Reality -- A Practitioner Systematization of Autonomous-Agent Threats and Controls in Regulated Financial Systems","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2606.10749","citing_title":"Toward Secure LLM Agents: Threat Surfaces, Attacks, Defenses, and Evaluation","ref_index":130,"is_internal_anchor":false},{"citing_arxiv_id":"2605.23330","citing_title":"Security, Privacy, and Ethical Risks in OpenClaw","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2605.22842","citing_title":"The Misattribution Gap: When Memory Poisoning Looks Like Model Failure in Agentic AI Systems","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2605.01970","citing_title":"Trojan Hippo: Weaponizing Agent Memory for Data Exfiltration","ref_index":51,"is_internal_anchor":false},{"citing_arxiv_id":"2603.09002","citing_title":"Security Considerations for Multi-agent Systems","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08468","citing_title":"PYTHALAB-MERA: Validation-Grounded Memory, Retrieval, and Acceptance Control for Frozen-LLM Coding Agents","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09721","citing_title":"Security Risks in Tool-Enabled AI Agents: A Systematic Analysis of Privileged Execution Environments","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2604.12986","citing_title":"Parallax: Why AI Agents That Think Must Never Act","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2605.01970","citing_title":"Trojan Hippo: Weaponizing Agent Memory for Data Exfiltration","ref_index":52,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JBZP6UTXLODYEDYXNRTN5LC5IZ","json":"https://pith.science/pith/JBZP6UTXLODYEDYXNRTN5LC5IZ.json","graph_json":"https://pith.science/api/pith-number/JBZP6UTXLODYEDYXNRTN5LC5IZ/graph.json","events_json":"https://pith.science/api/pith-number/JBZP6UTXLODYEDYXNRTN5LC5IZ/events.json","paper":"https://pith.science/paper/JBZP6UTX"},"agent_actions":{"view_html":"https://pith.science/pith/JBZP6UTXLODYEDYXNRTN5LC5IZ","download_json":"https://pith.science/pith/JBZP6UTXLODYEDYXNRTN5LC5IZ.json","view_paper":"https://pith.science/paper/JBZP6UTX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.19956&json=true","fetch_graph":"https://pith.science/api/pith-number/JBZP6UTXLODYEDYXNRTN5LC5IZ/graph.json","fetch_events":"https://pith.science/api/pith-number/JBZP6UTXLODYEDYXNRTN5LC5IZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JBZP6UTXLODYEDYXNRTN5LC5IZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JBZP6UTXLODYEDYXNRTN5LC5IZ/action/storage_attestation","attest_author":"https://pith.science/pith/JBZP6UTXLODYEDYXNRTN5LC5IZ/action/author_attestation","sign_citation":"https://pith.science/pith/JBZP6UTXLODYEDYXNRTN5LC5IZ/action/citation_signature","submit_replication":"https://pith.science/pith/JBZP6UTXLODYEDYXNRTN5LC5IZ/action/replication_record"}},"created_at":"2026-07-05T10:57:57.284935+00:00","updated_at":"2026-07-05T10:57:57.284935+00:00"}