The reading library

FastAPI tutorial, request bodies, dependencies, async, and Pydantic model validation.

Access-control, validation, authentication, file-handling, and logging requirements.

BOLA (API1), Broken Auth (API2), Unrestricted Resource Consumption (API4), SSRF (API7), Misconfiguration (API8).

Prompt injection, sensitive info disclosure, improper output handling, excessive agency, vector weaknesses, unbounded consumption.

ML attack tactics/techniques: prompt injection, data poisoning, model extraction, and exfiltration.

API endpoints, model management, Modelfiles, and local runtime network binding security.

Core schema types, required properties, additionalProperties: false, enums, and numeric/string bounds.

EventSource interface, event stream wire format, connection lifecycle, cancellation, and reconnection.

Foreign-key constraints, indexes, ACID transactions, row-level security (RLS), and query planning.

Authorization, SQL Injection Prevention, File Upload, SSRF, Secrets Management, and Logging Cheat Sheets.

Model cards, tokenizer/config files, embedding usage, and repository commit-hash revision pinning.

Zero-copy safe tensor format overview vs. unsafe pickle deserialization risks.

Collections, payload metadata indexing, pre-filtering, multitenancy partitioning, and snapshots.

RAG evaluation metrics: context precision, context recall, faithfulness, and answer relevance.

Retrieval evaluation methodology (nDCG@k, Recall@k) and why a single metric is insufficient.

Analyzer, anonymizer, custom PII recognizers, and probabilistic detection false-positive/negative limits.

Identify-P, Govern-P, Control-P, Communicate-P, and Protect-P lifecycle controls.

Differential privacy PrivacyEngine, gradient clipping, noise multiplier, $(\epsilon, \delta)$ accounting, and utility trade-offs.

Dataset loading, revision pinning, dataset cards, and data provenance tracking.

Model Cards for Model Reporting (Mitchell et al.): intended use, limitations, training data, and disaggregated evals.

Official specification: client/server lifecycle, JSON-RPC messages, tool discovery, authorization, and trust boundaries.

RFC 6749 core flows + RFC 9700 security best practices: narrow scopes, audience restriction, token leakage, and replay.

Delegated authorization patterns, sub (subject) vs. act (actor) claims for agent delegation.

State graphs, durable checkpointing, human-in-the-loop interrupts, retries, and resumability.

Daemon surface, Linux namespaces/capabilities, rootless/non-root containers, read-only FS, and secrets.

Serving architecture, OpenAI-compatible server, authentication, concurrency flags, and production deployment.

Traces, spans, context propagation, semantic conventions, and redacting sensitive attributes.

Python Risk Identification Tool: targets, orchestrators, converters, scorers, and multi-turn attack automation.

LLM vulnerability scanner architecture: probes, generators, detectors, and buffs (authorized targets only).

Improper output handling (LLM05): pairing schema validation with sink-specific encoding (XSS/SQLi/SSRF prevention).

Govern, Map, Measure, and Manage functions; risk ownership, measurement, and lifecycle documentation.

Risk-based regulatory tiers, high-risk system obligations, and GPAI transparency/cybersecurity rules.

Isolated BrowserContext, downloads, permissions, screenshots, bounding boxes, locators, and tracing.

HookedTransformer, residual stream, ActivationCache, Logit Lens, and Activation Patching tutorials.

CVE records, CPE affected-version ranges, CVSS v3/v4 scoring context, and Known Exploited Vulnerabilities catalog.

CycloneDX/SPDX SBOM generation, container image scanning, dependency vulnerability detection, and scanner limits.

Supply-chain Levels for Software Artifacts (SLSA), build provenance attestations, and Cosign artifact signing.

JSON Canonicalization Scheme: deterministic key sorting and number formatting required before cryptographic hashing.

Four-question framework & STRIDE: assets, trust boundaries, data flows, threat identification, and mitigation tracking.

Web Security Testing Guide: hands-on test cases for authentication, authorization, session, input, and API security.

Agentic threat taxonomy: tool misuse, excessive agency, memory poisoning, inter-agent trust, and human oversight.

NIST AI 600-1: generative AI risks and suggested actions for confabulation, privacy, information integrity, and value chain.

Automated supply-chain checks: pinned dependencies, least-privilege CI tokens, branch protection, and signed releases.

Software Component Verification Standard: component inventory, provenance, and supplier assurance practices.

Adversarial Machine Learning: A Taxonomy and Terminology of Attacks and Mitigations (evasion, poisoning, privacy, abuse).