@blamejs/exceptd-skills 0.13.93 → 0.13.95

package/CHANGELOG.md

@@ -1,5 +1,13 @@
 # Changelog
 
+## 0.13.95 — 2026-05-25
+
+CVE catalog — LlamaIndex CLI command injection. Adds **CVE-2025-1753** (CWE-78, CNA huntr.dev CVSS 7.8; NVD has not assigned its own score): the LlamaIndex CLI builds a shell command from the user-supplied `--files` argument and runs it without neutralization, so shell metacharacters execute arbitrary OS commands; the fix adds shlex escaping. Maps ATT&CK T1059, with a zero-day lesson (NEW-CTRL-100) requiring AI-framework CLIs/tools to use argv-array execution or shlex neutralization rather than building shell strings from arguments — the same root cause as the MCP-stdio command-injection family, applied to a framework CLI. CVE count 369 → 370.
+
+## 0.13.94 — 2026-05-25
+
+CVE catalog — AnythingLLM upload path traversal to RCE. Adds **CVE-2024-13059** (CWE-22, NIST CVSS 7.2): AnythingLLM's multer-based upload handler mishandles non-ASCII filenames so they decode into `../` traversal sequences, letting a manager/admin user write attacker content to an arbitrary path (e.g. a startup script) and achieve remote code execution on the host; fixed in 1.3.1. Maps ATLAS AML.T0049 and ATT&CK T1190 / T1059, and reuses the runtime-API path-traversal control (NEW-CTRL-094) shared with the Ollama entries — AI-app file/path inputs must be canonicalized and validated, including non-ASCII transforms, before touching the filesystem. CVE count 368 → 369.
+
 ## 0.13.93 — 2026-05-25
 
 CVE catalog — LangChain experimental-chain code execution (prompt injection to RCE). Adds the canonical class where an LLM chain turns prompt-influenced input into executed Python. **CVE-2024-21513** (langchain-experimental, CWE-94, NIST CVSS 8.5) — VectorSQLDatabaseChain evaluates database values as code, so an attacker controlling the input prompt achieves arbitrary code execution; fixed in 0.0.21. **CVE-2023-44467** (langchain_experimental PALChain, CWE-94, NIST CVSS 9.8) — PALChain executes prompt-generated Python and did not block the dunder-import builtin, bypassing the earlier CVE-2023-36258 fix; fixed in 0.0.306. Both map ATLAS AML.T0051 (LLM prompt injection) + AML.T0011 and ATT&CK T1059 / T1059.006, and their shared zero-day lesson (NEW-CTRL-099) requires chains that execute generated code to sandbox or disable it — builtin denylists are an incomplete fix. Distinct from the existing LangChain entries (LangGrinch serialization, Chatchat MCP). CVE count 366 → 368.

package/data/_indexes/_meta.json

@@ -1,21 +1,21 @@
 {
   "schema_version": "1.1.0",
-  "generated_at": "2026-05-26T02:33:31.075Z",
+  "generated_at": "2026-05-26T03:30:30.490Z",
   "generator": "scripts/build-indexes.js",
   "source_count": 54,
   "source_hashes": {
-    "manifest.json": "cb92f7a7f61228eb5632fbaeaa386b953c90bc2016eb9f84a7f2bb9f8f3013b1",
-    "data/atlas-ttps.json": "e3d139160b51eac0360f4bee92be52e77e0e441f8596b32f04e8bd6f29c1acd9",
-    "data/attack-techniques.json": "4f269298532ae7ef304b2f5e52dbb72e8d363e0f1d29dca7c691a7f15d23b7e5",
-    "data/cve-catalog.json": "0fae8ec235920068e1ea3bed6f3280e920feb2aa82bcb185a8dee42f419c3242",
-    "data/cwe-catalog.json": "1f2d77ee7d83e6aa113055f620a1c56d9287392c8c0a3fb8336abc52ce53d8fd",
+    "manifest.json": "ca9e058f24dff42048d0dbd2bc1cbe43d1bb6952a725ae3cd2b1d008d600d3ec",
+    "data/atlas-ttps.json": "00c4cd73790a452917a61423c12ecdaf4a7dd27544a9239665fbef4779ba0c6a",
+    "data/attack-techniques.json": "b3f989c348815c79f76a23ce147222e64e92e405f73fdfb60504691c05462b38",
+    "data/cve-catalog.json": "726ebd2d7428346544d033f7e05b10c9ca527d6fea545accb23b3646a9209760",
+    "data/cwe-catalog.json": "5298d58103a45efe5149c96f578416fe49f517da84a69560063f6dd7ff0dbca4",
     "data/d3fend-catalog.json": "9a54bccb9f24f84b32024216cc3f53819a053721ac8ab43c326859e68fc0ffaf",
     "data/dlp-controls.json": "d2406c482dddd30e49203879999dc4b3a7fd4d0494d6a61d86b91ee76415df19",
     "data/exploit-availability.json": "ec2656f0d9a893610e27b43eb6035fe9b18e057c9f6dfaac7e7d4959bbcbb795",
-    "data/framework-control-gaps.json": "818c34517b1612cf391379898ae3da6fba16523af1d15df61f7c82823ce5f338",
+    "data/framework-control-gaps.json": "e2db0d9fa8e439ff79f8e65ec66fc260cbc83ca2b22ef4455c6a551aef8c14fe",
     "data/global-frameworks.json": "9ba563a85f7f8d6c3c957de64945e20925a89d0ed6ea6fc561cf093811acf558",
     "data/rfc-references.json": "66ef2e1f444a2cf0c2700a754f0a66030bb8a91d9e68394b9537ea1fe8b904fe",
-    "data/zeroday-lessons.json": "20a009550c9f4b9d4cbb79aaec1883cc2db4cc0353238f4c8583aa1f1a622e9b",
+    "data/zeroday-lessons.json": "1075c96dff51cc9834ebf1c013203e0ad5a81e09366e10fe16a4a5aa4ca77e46",
     "skills/kernel-lpe-triage/skill.md": "08b3e9815ba481c57c80f5fc0ccbf5bb7cbb41f570c235ba6ff9596b8c07354d",
     "skills/ai-attack-surface/skill.md": "c4c1eb22a38ca7a959b5725222bab8fbd4f4044a548a93f3e288e6f698334b72",
     "skills/mcp-agent-trust/skill.md": "89ac89084391d2341b6513fefb1be2d36b93de1c130f057696219c1c59440f13",
@@ -72,7 +72,7 @@
       "dlp_refs": 0
     },
     "trigger_table_entries": 538,
-    "chains_cve_entries": 357,
+    "chains_cve_entries": 359,
     "chains_cwe_entries": 171,
     "jurisdictions_indexed": 29,
     "handoff_dag_nodes": 42,

package/data/_indexes/activity-feed.json

@@ -149,7 +149,7 @@
       "artifact": "data/cve-catalog.json",
       "path": "data/cve-catalog.json",
       "schema_version": "1.0.0",
-      "entry_count": 368
+      "entry_count": 370
     },
     {
       "date": "2026-05-18",
@@ -165,7 +165,7 @@
       "artifact": "data/zeroday-lessons.json",
       "path": "data/zeroday-lessons.json",
       "schema_version": "1.1.0",
-      "entry_count": 363
+      "entry_count": 365
     },
     {
       "date": "2026-05-17",

package/data/_indexes/catalog-summaries.json

@@ -62,7 +62,7 @@
         "rebuild_after_days": 365,
         "note": "Per-entry last_verified governs decay. Skills depending on this catalog must check entry freshness before high-stakes use."
       },
-      "entry_count": 368,
+      "entry_count": 370,
       "sample_keys": [
         "CVE-2025-53773",
         "CVE-2026-30615",
@@ -238,7 +238,7 @@
         "rebuild_after_days": 365,
         "note": "Per-entry last_verified governs decay. Skills depending on this catalog must check entry freshness before high-stakes use."
       },
-      "entry_count": 363,
+      "entry_count": 365,
       "sample_keys": [
         "CVE-2026-31431",
         "CVE-2025-53773",