mirror of
https://github.com/mukul975/Anthropic-Cybersecurity-Skills.git
synced 2026-06-26 19:54:37 +03:00
mukul975
8cae0648ec
Demand-driven expansion targeting the fastest-growing 2025-2026 threat and
skills categories (ISC2/WEF/CrowdStrike/Mandiant signals):
- AI Security (NEW domain, 12 skills): LLM red-teaming with garak/PyRIT,
prompt injection (direct/indirect/RAG), MCP tool-poisoning, agentic tool
invocation, guardrails, model/data poisoning, system-prompt leakage,
embedding/vector weaknesses, model extraction, continuous red-teaming
- Supply Chain Security (NEW domain, 5 skills): SBOMs, dependency confusion,
malicious-npm triage, typosquatting, SLSA/Sigstore provenance
- Hardware & Firmware Security (NEW domain, 4 skills): CHIPSEC/UEFI audit,
Secure Boot bypass, TPM measured-boot attestation, ESP bootkit hunting
- Identity (10): Entra ID/ROADtools, GraphRunner, AADInternals, ADCS/Certipy,
shadow credentials, coercion, BloodHound CE, device-code phishing, SSO abuse
- Cloud-native (8): Stratus, Pacu, CloudFox, container escape, K8s RBAC,
Falco, Trivy, kube-bench
- Offensive C2 (6): Sliver, Havoc, NetExec, DPAPI, NTLM relay ESC8, redirectors
- DFIR (6): Hayabusa, Chainsaw, KAPE, Velociraptor, EZ Tools, Plaso
- Backfill (4): OpenCTI, MISP, honeytokens, post-quantum crypto migration
Each skill follows the repo taxonomy (SKILL.md + references/{standards,api-reference}.md
+ scripts/agent.py + LICENSE), with researched real tool commands (no placeholders),
complete frontmatter, and ATT&CK/ATLAS + NIST CSF mappings. Updates README domain
table, skill count, and index.json.
53 lines
2.3 KiB
Markdown
53 lines
2.3 KiB
Markdown
# Model Extraction Detection — API / Library Reference
|
|
|
|
## Libraries
|
|
|
|
| Library | Install | Purpose |
|
|
|---------|---------|---------|
|
|
| adversarial-robustness-toolbox | `pip install adversarial-robustness-toolbox` | Extraction, inversion, and membership-inference attacks + defenses |
|
|
| scikit-learn | `pip install scikit-learn` | Surrogate / attack model training |
|
|
| numpy | `pip install numpy` | Confidence-vector math, perturbation |
|
|
|
|
## ART Extraction Attacks (`art.attacks.extraction`)
|
|
|
|
| Class | Key params | Purpose |
|
|
|-------|-----------|---------|
|
|
| `KnockoffNets` | `nb_stolen`, `batch_size_query`, `nb_epochs`, `sampling_strategy` | Train surrogate from black-box queries (Knockoff Nets) |
|
|
| `CopycatCNN` | `nb_stolen`, `batch_size_fit`, `batch_size_query` | Copycat surrogate extraction for neural nets |
|
|
| `attack.extract(x, thief_classifier=...)` | — | Run extraction; returns trained surrogate classifier |
|
|
|
|
## ART Inference Attacks (`art.attacks.inference.membership_inference`)
|
|
|
|
| Class | Key methods | Purpose |
|
|
|-------|-------------|---------|
|
|
| `MembershipInferenceBlackBox` | `.fit(...)`, `.infer(x, y)` | Black-box membership inference (AML.T0024.000) |
|
|
| `MembershipInferenceBlackBoxRuleBased` | `.infer(x, y)` | Rule-based MIA baseline (no shadow training) |
|
|
|
|
## ART Defenses (postprocessors)
|
|
|
|
| Class | Purpose |
|
|
|-------|---------|
|
|
| `art.defences.postprocessor.ReverseSigmoid` | Perturb output probabilities to hinder extraction |
|
|
| `art.defences.postprocessor.Rounded` | Round confidence values to reduce leaked precision |
|
|
| `art.defences.postprocessor.HighConfidence` | Suppress low-confidence outputs |
|
|
|
|
## Estimator Wrappers
|
|
|
|
| Class | Purpose |
|
|
|-------|---------|
|
|
| `art.estimators.classification.SklearnClassifier` | Wrap a scikit-learn model as an ART victim |
|
|
| `art.estimators.classification.KerasClassifier` / `PyTorchClassifier` | Wrap DL models |
|
|
|
|
## Detection Signals (custom)
|
|
|
|
| Signal | Heuristic |
|
|
|--------|-----------|
|
|
| Query volume | Queries/principal/window above baseline |
|
|
| Unique-input ratio | `unique(input_hash)/queries` → ~1.0 |
|
|
| Confidence-request ratio | Fraction of calls demanding full probability vectors |
|
|
|
|
## External References
|
|
|
|
- ART docs: https://adversarial-robustness-toolbox.readthedocs.io/
|
|
- MITRE ATLAS AML.T0024: https://atlas.mitre.org/techniques/AML.T0024
|