agent-threat-rules 2.0.11 → 2.0.12
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.
- package/README.md +9 -8
- package/package.json +2 -2
package/README.md
CHANGED
|
@@ -12,7 +12,7 @@ AI Agent 威脅偵測規則 -- 開源、社群驅動
|
|
|
12
12
|
[](https://pypi.org/project/pyatr/)
|
|
13
13
|
[](https://github.com/marketplace/actions/atr-scan)
|
|
14
14
|
[](LICENSE)
|
|
15
|
-
[](#what-atr-detects)
|
|
16
16
|
[](#ecosystem)
|
|
17
17
|
[](#evaluation)
|
|
18
18
|
[](#ecosystem-scan)
|
|
@@ -84,7 +84,7 @@ npm install -g agent-threat-rules
|
|
|
84
84
|
atr scan skill.md # scan a SKILL.md for threats
|
|
85
85
|
atr scan mcp-config.json # scan MCP events for threats
|
|
86
86
|
atr scan skill.md --sarif # output SARIF v2.1.0 for GitHub Security tab
|
|
87
|
-
atr convert generic-regex # export
|
|
87
|
+
atr convert generic-regex # export 311 rules as JSON (1,600+ regex patterns)
|
|
88
88
|
atr convert splunk # export to Splunk SPL
|
|
89
89
|
atr convert elastic # export to Elasticsearch Query DSL
|
|
90
90
|
atr stats # show rule collection stats
|
|
@@ -110,7 +110,7 @@ One line. Zero config. SARIF results in your Security tab.
|
|
|
110
110
|
|
|
111
111
|
## What ATR Detects
|
|
112
112
|
|
|
113
|
-
|
|
113
|
+
311 rules across 9 categories, mapped to real CVEs:
|
|
114
114
|
|
|
115
115
|
| Category | What it catches | Rules | Real CVEs |
|
|
116
116
|
|----------|----------------|-------|-----------|
|
|
@@ -259,7 +259,7 @@ Every rule is a YAML file answering: **what** to detect, **how** to detect it, *
|
|
|
259
259
|
### Export rules
|
|
260
260
|
|
|
261
261
|
```bash
|
|
262
|
-
# For your security platform (
|
|
262
|
+
# For your security platform (311 rules, 1,600+ regex patterns as JSON)
|
|
263
263
|
atr convert generic-regex --output atr-rules.json
|
|
264
264
|
|
|
265
265
|
# For SIEM integration
|
|
@@ -307,7 +307,7 @@ Want to integrate ATR into your product? Three options:
|
|
|
307
307
|
```bash
|
|
308
308
|
# Option 1: Export rules as JSON (recommended for most tools)
|
|
309
309
|
atr convert generic-regex --output atr-rules.json
|
|
310
|
-
# →
|
|
310
|
+
# → 311 rules, 1,600+ regex patterns, severity/category metadata
|
|
311
311
|
|
|
312
312
|
# Option 2: Use the TypeScript engine directly
|
|
313
313
|
npm install agent-threat-rules
|
|
@@ -358,7 +358,8 @@ See [CONTRIBUTING.md](CONTRIBUTING.md) for the full guide. See [CONTRIBUTION-GUI
|
|
|
358
358
|
- [x] **v0.4** -- 71 rules, ClawHub 36K scan, SAFE-MCP 91.8%
|
|
359
359
|
- [x] **v1.0** -- 108 rules, 53K mega scan, GitHub Action + SARIF, generic-regex export, Cisco adoption
|
|
360
360
|
- [x] **v1.1** -- Threat Cloud flywheel, 5 ecosystem merges, Microsoft AGT + NVIDIA Garak PRs
|
|
361
|
-
- [x] **v2.0.0**
|
|
361
|
+
- [x] **v2.0.0** -- 113 rules, 96K mega scan, 751 malware discovered, RFC-001, GOVERNANCE.md, website launch
|
|
362
|
+
- [x] **v2.0.11** (current) -- 311 rules, 193 new NVIDIA garak probe coverage (ATR-00300~00414), 97.1% garak recall
|
|
362
363
|
- [ ] **v2.1** -- Go engine, ML classifier integration, semantic signatures, community rule submissions
|
|
363
364
|
- [ ] **v3.0** -- Multi-engine standard: 2+ engines, 10+ production deployments, schema review by 3+ security teams
|
|
364
365
|
|
|
@@ -366,7 +367,7 @@ See [CONTRIBUTING.md](CONTRIBUTING.md) for the full guide. See [CONTRIBUTION-GUI
|
|
|
366
367
|
|
|
367
368
|
| Phase | Goal | Status |
|
|
368
369
|
|-------|------|--------|
|
|
369
|
-
| **Phase 0: Core product** |
|
|
370
|
+
| **Phase 0: Core product** | 311 rules, 97.1% garak recall, OWASP 10/10, 96K scan | **Done** |
|
|
370
371
|
| **Phase 1: Distribution** | GitHub Action, SARIF, generic-regex export, ecosystem PRs | **Done** |
|
|
371
372
|
| **Phase 2: Adoption** | Cisco merged (34 rules), OWASP PR, 11 ecosystem PRs | **In progress** |
|
|
372
373
|
| **Phase 3: Community flywheel** | Threat Cloud crystallization, auto-generated rules, 10+ contributors | In progress |
|
|
@@ -381,7 +382,7 @@ ATR uses "ATR Scanned" (not "ATR Certified") until recall exceeds 80%. We are ho
|
|
|
381
382
|
```
|
|
382
383
|
ATR (this repo) Your Product / Integration
|
|
383
384
|
┌─────────────────────────┐ ┌──────────────────────────┐
|
|
384
|
-
│
|
|
385
|
+
│ 311 Rules (YAML) │ match │ Block / Allow / Alert │
|
|
385
386
|
│ Engine (TS + Py) │ ────────→ │ SIEM (Splunk / Elastic) │
|
|
386
387
|
│ CLI / MCP / GitHub Act. │ results │ CI/CD (SARIF → Security) │
|
|
387
388
|
│ SARIF / Generic Regex │ │ Runtime Proxy (MCP) │
|
package/package.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "agent-threat-rules",
|
|
3
|
-
"version": "2.0.
|
|
3
|
+
"version": "2.0.12",
|
|
4
4
|
"type": "module",
|
|
5
|
-
"description": "Open detection standard for AI
|
|
5
|
+
"description": "Open detection standard -- like Sigma, but for AI agents. 311 rules for prompt injection, tool poisoning, context exfiltration, and MCP attacks. Shipped in Cisco AI Defense. 97.1% recall on NVIDIA garak.",
|
|
6
6
|
"main": "./dist/index.js",
|
|
7
7
|
"types": "./dist/index.d.ts",
|
|
8
8
|
"bin": {
|