結果RST-2026-0202v0.1
F1–F4 合併:每種表面都有高警示——README 3.4%、agent 指示檔 3.0%、治理檔 0.9%
主規則,236 個倉庫:README 8/233、agent 指示檔 7/236、治理檔 2/235。各輪從 1 格到 9 格。企業樣本「集中在 agent 指示檔」的現象沒有重現;具體路徑類(L3)仍偏向指示檔。
觀察到的結果
metricsby_surfaceGENERAL_READMEhigh_warning- 8
analyzable- 233
share- 0.0343
ci95- 0.0149
- 0.0665
l3- 2
AI_INSTRUCTION_SELECTEDhigh_warning- 7
analyzable- 236
share- 0.0297
ci95- 0.012
- 0.0602
l3- 4
GOVERNANCE_FAMILYhigh_warning- 2
analyzable- 235
share- 0.0085
ci95- 0.001
- 0.0304
l3- 1
pairedrepositories- 236
fully_analyzable- 232
high_warning_patternAI_INSTRUCTION_SELECTED- 5
GENERAL_README- 7
GENERAL_README+AI_INSTRUCTION_SELECTED- 1
GOVERNANCE_FAMILY- 2
NONE- 217
instruction_vs_readmeonly_AI_INSTRUCTION_SELECTED- 5
only_GENERAL_README- 7
instruction_vs_governanceonly_AI_INSTRUCTION_SELECTED- 6
only_GOVERNANCE_FAMILY- 2
repositories_by_owner_typeOrganization- 70
User- 166
詮釋
interpretation- High-warning context exists on every surface; surfaces are not separated by the intervals; any single round is unrepresentative.
限制
limitations- Exploratory, non-probability frame; not a prevalence estimate.
- No human checked any finding; secret-like means a shape and was never tested.
- Current public state only, observed once.
主張
| 來源 | 關係 | 目標 | 狀態 | ID |
|---|---|---|---|---|
RST-2026-0202 F1–F4 合併:每種表面都有高警示——README 3.4%、agent 指示檔 3.0%、治理檔 0.9% | supports | OBS-2026-0208 O8——與 agent 相關的公開邊界訊號,同樣出現在個人、組織與企業 | ACTIVE | REL-2026-0594 |
RST-2026-0202 F1–F4 合併:每種表面都有高警示——README 3.4%、agent 指示檔 3.0%、治理檔 0.9% | qualifies | OBS-2026-0205 O5——在凍結的 60 個企業倉庫樣本中,兩個高警示結果都在 agent 指示檔上 | ACTIVE | REL-2026-0595 |
關係
| 來源 | 關係 | 目標 | 狀態 | ID |
|---|---|---|---|---|
RST-2026-0202 F1–F4 合併:每種表面都有高警示——README 3.4%、agent 指示檔 3.0%、治理檔 0.9% | supports | OBS-2026-0208 O8——與 agent 相關的公開邊界訊號,同樣出現在個人、組織與企業 | ACTIVE | REL-2026-0594 |
RST-2026-0202 F1–F4 合併:每種表面都有高警示——README 3.4%、agent 指示檔 3.0%、治理檔 0.9% | qualifies | OBS-2026-0205 O5——在凍結的 60 個企業倉庫樣本中,兩個高警示結果都在 agent 指示檔上 | ACTIVE | REL-2026-0595 |
EXP-2026-0202 F1–F4 輪——表面平衡的新樣本(236 個倉庫,2026-10-13) | produces | RST-2026-0202 F1–F4 合併:每種表面都有高警示——README 3.4%、agent 指示檔 3.0%、治理檔 0.9% | ACTIVE | REL-2026-0593 |
歷史與來源歷程
- Canonical URL
- https://evemisslab.com/ai/results/RST-2026-0202/
- 快照
AI-SNAPSHOT-v0.1-58e716495670- 來源歷程
source- EveMissLab research collection: AI Agent Publication Boundary and Security Closure (真本體論14)
extracted_by- Splice (Claude Code), reading the open-source release's own manifests, locks and pooled outputs, and the hashes of the internal drafts
extracted_at- 2026-10-13
generator- tools/extract_all.py
claim_boundary- status, evidence level and result type follow the source artifact's own stated claim boundary; nothing is upgraded beyond what the report supports