AgentPROV corpus: AgentHarm runs of the agent security tools survey

An execution-provenance corpus built by AgentPROV v0.1.0 from the Inspect AgentHarm logs of the agent security tools survey (CC BY 4.0; the AgentHarm data keep their use clause). This deposit draft holds 1 AgentPROV corpus version(s), prepared offline by agentprov package --deposit for the release date 2026-10-30. Nothing has been uploaded or published. Corpus ast-survey-agentharm 0.1.0: Agent Security Tools Survey: AgentHarm Inspect logs (restricted-use corpus). 1055 execution records, of which 0 are outcome-only records. Licence class: CC-BY-4.0; restrictions: "AgentHarm task data: MIT License with an additional clause (use only to improve AI safety and security)". Content digest sha256:c9b76d5e3b590bae3edf8a4cf3a617a90e9fd07b1447373b02bd420327b5319a; root digest sha256:ba9c2206fc207f0ff09653e353c754a5a758e257d8c6cf54db8515c7bcf9f229. Counts are computed from each corpus manifest at sealing. Each DATASHEET states what the records do and do not establish; a statement about the data is scoped as it is there. To verify: compare each archive's sha256 or md5 with DEPOSIT-MANIFEST.json, run agentprov verify on the archives (archive mode, no extraction; H.5.5), and compare each root digest with the one published outside the deposit (the corpus tag annotation and the paper).