agent-threat-rules 4.1.5 → 4.2.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (240) hide show
  1. package/CHANGELOG.md +688 -1
  2. package/README.md +355 -125
  3. package/dist/adapters/openshell-filter.d.ts +10 -3
  4. package/dist/adapters/openshell-filter.d.ts.map +1 -1
  5. package/dist/adapters/openshell-filter.js +10 -3
  6. package/dist/adapters/openshell-filter.js.map +1 -1
  7. package/dist/adapters/stdio-adapter.js +3 -3
  8. package/dist/adapters/stdio-adapter.js.map +1 -1
  9. package/dist/cli/mcp-tools-document.d.ts +40 -0
  10. package/dist/cli/mcp-tools-document.d.ts.map +1 -0
  11. package/dist/cli/mcp-tools-document.js +95 -0
  12. package/dist/cli/mcp-tools-document.js.map +1 -0
  13. package/dist/cli/scan-handler.d.ts +8 -0
  14. package/dist/cli/scan-handler.d.ts.map +1 -1
  15. package/dist/cli/scan-handler.js +126 -34
  16. package/dist/cli/scan-handler.js.map +1 -1
  17. package/dist/cli.d.ts.map +1 -1
  18. package/dist/cli.js +80 -43
  19. package/dist/cli.js.map +1 -1
  20. package/dist/converters/sarif.d.ts +14 -2
  21. package/dist/converters/sarif.d.ts.map +1 -1
  22. package/dist/converters/sarif.js +7 -9
  23. package/dist/converters/sarif.js.map +1 -1
  24. package/dist/corpus-event.d.ts +47 -5
  25. package/dist/corpus-event.d.ts.map +1 -1
  26. package/dist/corpus-event.js +68 -29
  27. package/dist/corpus-event.js.map +1 -1
  28. package/dist/embedding/model-loader.d.ts +2 -0
  29. package/dist/embedding/model-loader.d.ts.map +1 -1
  30. package/dist/embedding/model-loader.js +3 -1
  31. package/dist/embedding/model-loader.js.map +1 -1
  32. package/dist/enforcement.d.ts +8 -0
  33. package/dist/enforcement.d.ts.map +1 -1
  34. package/dist/enforcement.js +10 -0
  35. package/dist/enforcement.js.map +1 -1
  36. package/dist/engine.d.ts +150 -6
  37. package/dist/engine.d.ts.map +1 -1
  38. package/dist/engine.js +337 -76
  39. package/dist/engine.js.map +1 -1
  40. package/dist/eval/embedding-provenance.d.ts +40 -0
  41. package/dist/eval/embedding-provenance.d.ts.map +1 -0
  42. package/dist/eval/embedding-provenance.js +71 -0
  43. package/dist/eval/embedding-provenance.js.map +1 -0
  44. package/dist/eval/eval-harness.d.ts +34 -0
  45. package/dist/eval/eval-harness.d.ts.map +1 -1
  46. package/dist/eval/eval-harness.js +49 -2
  47. package/dist/eval/eval-harness.js.map +1 -1
  48. package/dist/eval/hackaprompt-corpus.d.ts +12 -0
  49. package/dist/eval/hackaprompt-corpus.d.ts.map +1 -1
  50. package/dist/eval/hackaprompt-corpus.js +15 -1
  51. package/dist/eval/hackaprompt-corpus.js.map +1 -1
  52. package/dist/eval/hackaprompt-provenance.d.ts +52 -0
  53. package/dist/eval/hackaprompt-provenance.d.ts.map +1 -0
  54. package/dist/eval/hackaprompt-provenance.js +108 -0
  55. package/dist/eval/hackaprompt-provenance.js.map +1 -0
  56. package/dist/eval/run-hackaprompt-benchmark.d.ts +18 -8
  57. package/dist/eval/run-hackaprompt-benchmark.d.ts.map +1 -1
  58. package/dist/eval/run-hackaprompt-benchmark.js +154 -64
  59. package/dist/eval/run-hackaprompt-benchmark.js.map +1 -1
  60. package/dist/eval/run-pint-benchmark.js +11 -1
  61. package/dist/eval/run-pint-benchmark.js.map +1 -1
  62. package/dist/eval/skill-benchmark.d.ts +26 -6
  63. package/dist/eval/skill-benchmark.d.ts.map +1 -1
  64. package/dist/eval/skill-benchmark.js +49 -16
  65. package/dist/eval/skill-benchmark.js.map +1 -1
  66. package/dist/flywheel.d.ts +4 -0
  67. package/dist/flywheel.d.ts.map +1 -1
  68. package/dist/flywheel.js +12 -4
  69. package/dist/flywheel.js.map +1 -1
  70. package/dist/hook-handler.d.ts +11 -1
  71. package/dist/hook-handler.d.ts.map +1 -1
  72. package/dist/hook-handler.js +84 -16
  73. package/dist/hook-handler.js.map +1 -1
  74. package/dist/hook-tool-output.d.ts +47 -0
  75. package/dist/hook-tool-output.d.ts.map +1 -0
  76. package/dist/hook-tool-output.js +110 -0
  77. package/dist/hook-tool-output.js.map +1 -0
  78. package/dist/index.d.ts +5 -3
  79. package/dist/index.d.ts.map +1 -1
  80. package/dist/index.js +3 -2
  81. package/dist/index.js.map +1 -1
  82. package/dist/layer-integration.d.ts +4 -2
  83. package/dist/layer-integration.d.ts.map +1 -1
  84. package/dist/layer-integration.js +6 -2
  85. package/dist/layer-integration.js.map +1 -1
  86. package/dist/mcp-server.d.ts +8 -0
  87. package/dist/mcp-server.d.ts.map +1 -1
  88. package/dist/mcp-server.js +1 -0
  89. package/dist/mcp-server.js.map +1 -1
  90. package/dist/mcp-tools/scan.d.ts.map +1 -1
  91. package/dist/mcp-tools/scan.js +9 -1
  92. package/dist/mcp-tools/scan.js.map +1 -1
  93. package/dist/measurement/enforce-lane.d.ts +66 -0
  94. package/dist/measurement/enforce-lane.d.ts.map +1 -0
  95. package/dist/measurement/enforce-lane.js +89 -0
  96. package/dist/measurement/enforce-lane.js.map +1 -0
  97. package/dist/measurement/from-eval-harness.d.ts +16 -3
  98. package/dist/measurement/from-eval-harness.d.ts.map +1 -1
  99. package/dist/measurement/from-eval-harness.js +15 -5
  100. package/dist/measurement/from-eval-harness.js.map +1 -1
  101. package/dist/measurement/in-sample-ablation.d.ts +29 -0
  102. package/dist/measurement/in-sample-ablation.d.ts.map +1 -0
  103. package/dist/measurement/in-sample-ablation.js +45 -0
  104. package/dist/measurement/in-sample-ablation.js.map +1 -0
  105. package/dist/measurement/schema.d.ts +38 -1
  106. package/dist/measurement/schema.d.ts.map +1 -1
  107. package/dist/measurement/schema.js +43 -1
  108. package/dist/measurement/schema.js.map +1 -1
  109. package/dist/quality/action-eligibility.d.ts +4 -2
  110. package/dist/quality/action-eligibility.d.ts.map +1 -1
  111. package/dist/quality/action-eligibility.js.map +1 -1
  112. package/dist/rule-confidence.d.ts +31 -0
  113. package/dist/rule-confidence.d.ts.map +1 -0
  114. package/dist/rule-confidence.js +49 -0
  115. package/dist/rule-confidence.js.map +1 -0
  116. package/dist/rule-scaffolder.d.ts.map +1 -1
  117. package/dist/rule-scaffolder.js +43 -4
  118. package/dist/rule-scaffolder.js.map +1 -1
  119. package/dist/semantic-evaluator.d.ts.map +1 -1
  120. package/dist/semantic-evaluator.js +15 -13
  121. package/dist/semantic-evaluator.js.map +1 -1
  122. package/dist/shadow-evaluator.d.ts +7 -2
  123. package/dist/shadow-evaluator.d.ts.map +1 -1
  124. package/dist/shadow-evaluator.js +29 -25
  125. package/dist/shadow-evaluator.js.map +1 -1
  126. package/dist/testcase-event.d.ts +82 -1
  127. package/dist/testcase-event.d.ts.map +1 -1
  128. package/dist/testcase-event.js +135 -16
  129. package/dist/testcase-event.js.map +1 -1
  130. package/dist/tier0-invariant.d.ts.map +1 -1
  131. package/dist/tier0-invariant.js +1 -0
  132. package/dist/tier0-invariant.js.map +1 -1
  133. package/dist/tier1-blacklist.d.ts.map +1 -1
  134. package/dist/tier1-blacklist.js +1 -0
  135. package/dist/tier1-blacklist.js.map +1 -1
  136. package/dist/tool-definition-scope.d.ts +45 -0
  137. package/dist/tool-definition-scope.d.ts.map +1 -0
  138. package/dist/tool-definition-scope.js +68 -0
  139. package/dist/tool-definition-scope.js.map +1 -0
  140. package/dist/types.d.ts +42 -2
  141. package/dist/types.d.ts.map +1 -1
  142. package/dist/verdict.d.ts +46 -2
  143. package/dist/verdict.d.ts.map +1 -1
  144. package/dist/verdict.js +99 -14
  145. package/dist/verdict.js.map +1 -1
  146. package/package.json +2 -1
  147. package/rules/agent-manipulation/ATR-2026-00030-cross-agent-attack.yaml +2 -2
  148. package/rules/agent-manipulation/ATR-2026-00139-casual-authority-redirect.yaml +1 -1
  149. package/rules/agent-manipulation/ATR-2026-00328-named-mode-persona-override.yaml +1 -1
  150. package/rules/agent-manipulation/ATR-2026-00330-yoja-allcaps-settings-block-persona.yaml +1 -1
  151. package/rules/agent-manipulation/ATR-2026-00418-weknora-mcp-config-rce.yaml +5 -5
  152. package/rules/agent-manipulation/ATR-2026-00432-superagi-output-handler-eval-rce.yaml +1 -1
  153. package/rules/agent-manipulation/ATR-2026-00440-semantic-kernel-vector-store-eval-rce.yaml +3 -3
  154. package/rules/context-exfiltration/ATR-2026-00426-nl-output-injection-credential-leak.yaml +1 -1
  155. package/rules/context-exfiltration/ATR-2026-00702-ipi-credential-exfil-via-agent-action.yaml +1 -1
  156. package/rules/context-exfiltration/ATR-2026-00707-ipi-file-exfil-with-destroy-or-tamper.yaml +1 -1
  157. package/rules/context-exfiltration/ATR-2026-00851-indirect-pi-financial-payment-exfil.yaml +1 -1
  158. package/rules/context-exfiltration/ATR-2026-01457-sysprompt-completion-clone-attack.yaml +1 -1
  159. package/rules/context-exfiltration/ATR-2026-01462-sysprompt-override-pseudo-admin.yaml +1 -1
  160. package/rules/context-exfiltration/ATR-2026-01463-imend-imsystem-prompt-injection-format.yaml +1 -1
  161. package/rules/context-exfiltration/ATR-2026-01606-ssrf-internal-network-scan.yaml +1 -1
  162. package/rules/context-exfiltration/ATR-2026-02847-policy-justified-exfil-to-email.yaml +1 -1
  163. package/rules/context-exfiltration/ATR-2026-02855-cross-email-exfil-relay.yaml +79 -0
  164. package/rules/context-exfiltration/ATR-2026-02861-exfil-send-email-with-body-directive.yaml +83 -0
  165. package/rules/context-exfiltration/ATR-2026-02866-idna-label-separator-in-fetch-url-host.yaml +104 -0
  166. package/rules/excessive-autonomy/ATR-2026-00710-ipi-physical-access-grant.yaml +1 -1
  167. package/rules/model-abuse/ATR-2026-01897-host-sabotage.yaml +2 -2
  168. package/rules/model-security/ATR-2026-00433-modelcache-torch-load-deserialization-rce.yaml +1 -1
  169. package/rules/privilege-escalation/ATR-2026-00436-enclave-vm-sandbox-escape-rce.yaml +1 -1
  170. package/rules/privilege-escalation/ATR-2026-01601-sql-injection-destructive-ddl.yaml +2 -2
  171. package/rules/privilege-escalation/ATR-2026-01612-debug-mode-privilege-escalation.yaml +1 -1
  172. package/rules/privilege-escalation/ATR-2026-02867-identifier-typed-tool-argument-dot-dot-path-segment.yaml +106 -0
  173. package/rules/prompt-injection/ATR-2026-00004-system-prompt-override.yaml +2 -2
  174. package/rules/prompt-injection/ATR-2026-00080-encoding-evasion.yaml +1 -1
  175. package/rules/prompt-injection/ATR-2026-00086-visual-spoofing.yaml +1 -1
  176. package/rules/prompt-injection/ATR-2026-00203-context-pollution-skill-description.yaml +1 -1
  177. package/rules/prompt-injection/ATR-2026-00228-structured-jailbreak.yaml +2 -2
  178. package/rules/prompt-injection/ATR-2026-00233-structured-jailbreak.yaml +2 -2
  179. package/rules/prompt-injection/ATR-2026-00256-base-n-encoding-jailbreak.yaml +1 -1
  180. package/rules/prompt-injection/ATR-2026-00264-latent-injection-translation.yaml +3 -3
  181. package/rules/prompt-injection/ATR-2026-00265-latent-injection-rag-document.yaml +1 -1
  182. package/rules/prompt-injection/ATR-2026-00282-perez-prompt-injection-hijack.yaml +4 -4
  183. package/rules/prompt-injection/ATR-2026-00321-hyphenated-system-instruction-injection.yaml +1 -1
  184. package/rules/prompt-injection/ATR-2026-00394-backspace-deletion-character-injection.yaml +2 -2
  185. package/rules/prompt-injection/ATR-2026-00446-variable-assignment-payload-injection.yaml +3 -3
  186. package/rules/prompt-injection/ATR-2026-00476-garak-web-injection-javascript-alert-1.yaml +3 -3
  187. package/rules/prompt-injection/ATR-2026-00481-garak-dan-mixedunassigned.yaml +1 -1
  188. package/rules/prompt-injection/ATR-2026-00489-garak-encoding-javascript-alert-1.yaml +3 -3
  189. package/rules/prompt-injection/ATR-2026-00508-escape-delimiter-wrapped-goal-hijacking-in-user-input-prompt.yaml +3 -3
  190. package/rules/prompt-injection/ATR-2026-00512-rules-file-backdoor-injection.yaml +1 -1
  191. package/rules/prompt-injection/ATR-2026-01005-many-shot-repetition-override.yaml +2 -2
  192. package/rules/prompt-injection/ATR-2026-01006-payload-split-string-concat-injection.yaml +2 -2
  193. package/rules/prompt-injection/ATR-2026-01015-evasion-refusal-suppression.yaml +1 -1
  194. package/rules/prompt-injection/ATR-2026-01016-evasion-response-priming-forced-opener.yaml +1 -1
  195. package/rules/prompt-injection/ATR-2026-01026-evasion-unicode-glitch-token-anomaly.yaml +2 -2
  196. package/rules/prompt-injection/ATR-2026-01304-indirect-injection-carrier-important-override.yaml +1 -1
  197. package/rules/prompt-injection/ATR-2026-01759-dos-new-important-instructions-injection-in-document.yaml +1 -1
  198. package/rules/prompt-injection/ATR-2026-01865-llmail-fake-email-boundary-marker.yaml +1 -1
  199. package/rules/prompt-injection/ATR-2026-02010-emoji-only-payload.yaml +2 -2
  200. package/rules/prompt-injection/ATR-2026-02815-injected-halt-order-canned-reply-dos.yaml +80 -0
  201. package/rules/prompt-injection/ATR-2026-02820-tool-result-orders-agent-to-hide-its-own-action.yaml +106 -0
  202. package/rules/prompt-injection/ATR-2026-02858-context-suppression-directive.yaml +73 -0
  203. package/rules/prompt-injection/ATR-2026-02859-tool-output-agent-role-reassignment.yaml +71 -0
  204. package/rules/prompt-injection/ATR-2026-02860-forged-instruction-boundary-marker.yaml +70 -0
  205. package/rules/prompt-injection/ATR-2026-02862-fake-chat-template-turn-injection.yaml +79 -0
  206. package/rules/skill-compromise/ATR-2026-00060-skill-impersonation.yaml +1 -1
  207. package/rules/skill-compromise/ATR-2026-00063-skill-chain-attack.yaml +2 -2
  208. package/rules/skill-compromise/ATR-2026-00066-parameter-injection.yaml +2 -2
  209. package/rules/skill-compromise/ATR-2026-00149-skill-exfil-compound.yaml +6 -6
  210. package/rules/skill-compromise/ATR-2026-00223-reverse-shell-dropper.yaml +2 -2
  211. package/rules/skill-compromise/ATR-2026-01755-backdoor-pot-linguistic-trigger-phrase.yaml +1 -1
  212. package/rules/skill-compromise/ATR-2026-01756-backdoor-pot-symbol-emoticon-trigger.yaml +1 -1
  213. package/rules/skill-compromise/ATR-2026-02261-code-credential-exfil-chain.yaml +1 -1
  214. package/rules/tool-poisoning/ATR-2026-00012-unauthorized-tool-call.yaml +23 -23
  215. package/rules/tool-poisoning/ATR-2026-00013-tool-ssrf.yaml +16 -5
  216. package/rules/tool-poisoning/ATR-2026-00161-important-tag-cross-tool-shadowing.yaml +1 -1
  217. package/rules/tool-poisoning/ATR-2026-00277-echo-template-command-injection.yaml +2 -2
  218. package/rules/tool-poisoning/ATR-2026-00494-garak-exploitation-mixedunassigned.yaml +4 -4
  219. package/rules/tool-poisoning/ATR-2026-00581-mcp-tool-rug-pull-post-approval-redefinition.yaml +1 -1
  220. package/rules/tool-poisoning/ATR-2026-01303-tool-schema-enumeration-social-engineering.yaml +1 -1
  221. package/rules/tool-poisoning/ATR-2026-01307-mcp-dns-rebinding-attack.yaml +2 -2
  222. package/rules/tool-poisoning/ATR-2026-01775-semantic-mcp-tool-manifest-poisoning.yaml +1 -1
  223. package/rules/tool-poisoning/ATR-2026-01932-shadow-undeclared-mcp-server-registration.yaml +1 -1
  224. package/rules/tool-poisoning/ATR-2026-02025-mcp-full-schema-poisoning-nondescription-field.yaml +1 -1
  225. package/rules/tool-poisoning/ATR-2026-02817-untrusted-content-dictates-tool-arguments.yaml +83 -0
  226. package/rules/tool-poisoning/ATR-2026-02864-non-canonical-ipv4-host-in-url-tool-argument.yaml +102 -0
  227. package/rules/tool-poisoning/ATR-2026-02865-read-only-utility-write-flag-in-command-argument.yaml +157 -0
  228. package/rules/tool-poisoning/ATR-2026-02868-ssrf-ipv6-transition-address-embedded-private-ipv4.yaml +107 -0
  229. package/spec/README.md +6 -3
  230. package/spec/atr-event-v1.0.md +13 -8
  231. package/spec/atr-method-v1.1.md +2 -2
  232. package/spec/category-registry/gemara/README.md +102 -0
  233. package/spec/category-registry/gemara/ai-agent-capability-catalog.yaml +157 -0
  234. package/spec/category-registry/gemara/atr-attack-vector-catalog.yaml +130 -0
  235. package/spec/category-registry/gemara/atr-detection-control-catalog.yaml +327 -0
  236. package/spec/category-registry/gemara/atr-threat-catalog.yaml +246 -0
  237. package/spec/category-registry/gemara/src/editorial.yaml +743 -0
  238. package/spec/mappings/README.md +1 -1
  239. package/spec/mappings/atr-to-nist-csf-2.0.md +5 -4
  240. package/spec/schema/event.schema.json +1 -1
@@ -14,7 +14,7 @@ itself; the normative crosswalk fields live on individual Rules under
14
14
 
15
15
  | File | External framework | Status | Purpose |
16
16
  |------|--------------------|--------|---------|
17
- | [atr-to-nist-csf-2.0.md](./atr-to-nist-csf-2.0.md) | NIST CSF 2.0 (CSWP 29, Feb 2024) | Draft v1.0.0 | NIST IR 8596 Cyber AI Profile Informative Reference submission |
17
+ | [atr-to-nist-csf-2.0.md](./atr-to-nist-csf-2.0.md) | NIST CSF 2.0 (CSWP 29, Feb 2024) | Draft v1.0.0 | Draft; not submitted to NIST |
18
18
 
19
19
  Planned (not yet drafted):
20
20
 
@@ -1,7 +1,7 @@
1
1
  # ATR → NIST Cybersecurity Framework 2.0 Mapping
2
2
 
3
3
  Version: 1.1.0
4
- Status: Draft for NIST IR 8596 Informative Reference submission
4
+ Status: Draft. Not submitted to NIST.
5
5
  Date: 2026-06-14
6
6
  Editor: Adam Lin (林冠辛) <adam@agentthreatrule.org>
7
7
  Mapped corpus: Agent Threat Rules v3.5.0 (652 rules / 10 categories; per data/stats.json 2026-06-16)
@@ -12,8 +12,9 @@ Reference framework: NIST CSF 2.0 (NIST CSWP 29, February 2024)
12
12
  ## 1. Purpose
13
13
 
14
14
  This document maps the Agent Threat Rules (ATR) detection corpus to the NIST
15
- Cybersecurity Framework 2.0 subcategory taxonomy. It is intended as an
16
- Informative Reference submission per [NIST IR 8596 Cyber AI Profile][nist-ir-8596].
15
+ Cybersecurity Framework 2.0 subcategory taxonomy. It is written so it could
16
+ be offered as an Informative Reference under [NIST IR 8596 Cyber AI Profile][nist-ir-8596];
17
+ it has not been submitted.
17
18
 
18
19
  For each of the 10 ATR attack-class categories and each of the 5 detection
19
20
  methods (atr-method-v1.1.md §4), this mapping enumerates the CSF 2.0
@@ -173,7 +174,7 @@ attention:
173
174
  | Cross-tenant scope drift (invariant primitive) | GV.SC-04 (supplier criticality), GV.OC-04 (legal/regulatory requirements understood) |
174
175
  | Goal drift / autonomy escape (composite) | GV.RM-01 (risk management strategy), GV.SC-07 (risks from suppliers monitored) |
175
176
 
176
- ## 8. Implementation Guidance (for NIST IR 8596 reviewers)
177
+ ## 8. Implementation Guidance
177
178
 
178
179
  An organization seeking to use ATR as an Informative Reference for CSF 2.0
179
180
  implementation:
@@ -173,7 +173,7 @@
173
173
  },
174
174
  "agent.identity_assertion": {
175
175
  "type": ["string", "null"],
176
- "description": "JWT-format identity assertion per IETF AI agent auth drafts (when finalized)."
176
+ "description": "Reference to the identity assertion that authorised the agent, written <issuer>#<assertion id> (for a JWT, its iss and jti, or a SHA-256 digest of the token when it has no jti). Never the assertion itself: an assertion is a bearer credential, and raw tokens are forbidden in events."
177
177
  },
178
178
  "session.id": {
179
179
  "type": "string",