prodsec-skills

Security-focused review of AI-generated or AI-assisted code. Use when reviewing code produced by AI coding assistants, auditing AI-generated patches, verifying AI-assisted contributions before merge, or when a review needs to account for failure modes specific to AI code generation.

Configure Kubernetes network security including host access restrictions, network policies, port management, and connectivity requirements. Use when writing, reviewing, or auditing pod specs, Services, NetworkPolicies, or NetworkAttachmentDefinitions.

Secure AI-powered products and infrastructure. Use when integrating LLM APIs, deploying model-serving infrastructure, building agent workflows, connecting MCP servers, or reviewing AI system architecture for security gaps.

Enforce standard authentication for external data source connections. Use when integrating, configuring, or reviewing connections from AI systems to external databases, APIs, or data services.

Harden container images and runtime configuration. Use when building, reviewing, or auditing Containerfiles, Dockerfiles, container compose files, or Kubernetes pod security settings.

Supply chain security audit using CLI vulnerability scanners, license compliance checks, and supply chain risk assessment. Use when auditing project dependencies for known CVEs, license concerns, and supply chain health.

Apply when reviewing or writing code that processes untrusted input, constructs queries or commands, or handles user-supplied data. Covers SQL, LDAP, OS command injection, prototype pollution, and general validation strategy.

Enforce logging of access to sensitive external data sources in AI systems. Use when designing, building, or reviewing audit and logging capabilities for AI systems accessing external databases or data services.

Configure Kubernetes RBAC bindings, service accounts, namespaces, resource quotas, and service types for pods. Use when writing, reviewing, or auditing Deployments, ServiceAccounts, Roles, RoleBindings, ResourceQuotas, or Services for least-privilege access control.

Review OpenShift Security Context Constraints for correct privilege levels. Use when auditing pod security, reviewing SCC assignments, or configuring container runtime permissions on OpenShift clusters.

Detect hardcoded secrets in source code using pattern-based scanning for API keys, tokens, passwords, private keys, and connection strings. Use before committing, during code review, or for post-incident history scanning. Self-contained — no external tools required.

Apply Secure by Design, by Default, and in Deployment (SD3) principles. Use when reviewing default configurations, setup processes, update mechanisms, or evaluating whether a system fails open or closed.

Review web application security controls against OWASP-aligned risks. Use when building, auditing, or reviewing server-side web applications that handle user input, sessions, authentication, or access control.

Enables ultra-granular, line-by-line code analysis to build deep architectural context before vulnerability or bug finding.

Performs security-focused differential review of code changes (PRs, commits, diffs). Adapts analysis depth to codebase size, uses git history for context, calculates blast radius, checks test coverage, and generates comprehensive markdown reports. Automatically detects and prevents security regressions.

Run Semgrep static analysis scan on a codebase using parallel subagents. Supports two scan modes — "run all" (full ruleset coverage) and "important only" (high-confidence security vulnerabilities). Automatically detects and uses Semgrep Pro for cross-file taint analysis when available. Use when asked to scan code for vulnerabilities, run a security audit with Semgrep, find bugs, or perform static analysis. Spawns parallel workers for multi-language codebases.

Apply economy of mechanism and least common mechanism to reduce attack surface through simplicity and component isolation. Use when reviewing system architecture for unnecessary complexity, shared resources, or single points of failure.

Analyze OCI container images for security vulnerabilities, misconfigurations, supply chain risks, and hardening gaps. Use when reviewing container images from Red Hat, Quay.io, Docker Hub, or any OCI-compliant registry. Covers CVE analysis, config security, EOL component detection, credential exposure, and TLS/crypto misconfigurations.

Secure the administrative interface of model registries. Use when building, deploying, or reviewing model registry access controls and admin interfaces.

AFL++ is a fork of AFL with better fuzzing performance and advanced features. Use for multi-core fuzzing of C/C++ projects.

Audits GitHub Actions workflows for security vulnerabilities in AI agent integrations including Claude Code Action, Gemini CLI, OpenAI Codex, and GitHub AI Inference. Detects attack vectors where attacker-controlled input reaches AI agents running in CI/CD pipelines, including env var intermediary patterns, direct expression injection, dangerous sandbox configurations, and wildcard user allowlists. Use when reviewing workflow files that invoke AI coding agents, auditing CI/CD pipeline security for prompt injection risks, or evaluating agentic action configurations.

Enforce AWS security baselines for IAM, networking, logging, and managed services like RDS. Use when building, reviewing, or auditing AWS infrastructure, IAM policies, VPC configuration, or RDS deployments.

Apply when reviewing or writing GitLab CI (.gitlab-ci.yml), Tekton pipeline/task YAML, or Containerfiles/Dockerfiles used in CI builds. Detects common misconfigurations that expose secrets, weaken isolation, skip security gates, grant excessive privileges, or produce insecure container images.

cargo-fuzz is the de facto fuzzing tool for Rust projects using Cargo. Use for fuzzing Rust code with libFuzzer backend.

Apply when reviewing or writing browser-facing code that handles user input, renders HTML, sets cookies, or loads third-party scripts. Covers XSS, CSRF, CSP, clickjacking, XS-Leaks, and HTTP security headers.

Apply compiler hardening flags and sanitizers to C/C++ builds. Use when reviewing build configurations, CI pipelines, or Makefiles/CMakeLists for native code projects.

Coverage analysis measures code exercised during fuzzing. Use when assessing harness effectiveness or identifying fuzzing blockers.

Enforce database security for schema design, access control, encryption, and operational hardening. Use when building, reviewing, or auditing database schemas, connection handling, credential management, or database deployment configuration.

Enforce FIPS 140-2/140-3 compliance for RHEL, OpenShift, and Go workloads. Use when building, configuring, reviewing, or auditing systems that require FIPS-validated cryptographic modules, RHEL crypto-policy enforcement, FIPS-ready Go binaries, or FIPS-mode kernel and cluster configuration.

Systematically verifies suspected security bugs to eliminate false positives. Produces TRUE POSITIVE or FALSE POSITIVE verdicts with documented evidence for each bug.

Techniques for patching code to overcome fuzzing obstacles. Use when checksums, global state, or other barriers block fuzzer progress.

Harden locally deployed MCP servers against network exposure and host compromise. Use when building, deploying, or reviewing MCP servers that run on a user's local machine.

Techniques for writing effective fuzzing harnesses across languages. Use when creating new fuzz targets or improving existing harness code.

Verify and configure HTTP response security headers. Use when reviewing web server configuration, reverse proxy setup, or application middleware that sets response headers.

Enforce isolation and sandboxing of inference engine execution. Use when deploying, configuring, or reviewing inference engines like vLLM that load and serve AI models, especially third-party models.

Enforce least privilege and complete mediation in access control design. Use when reviewing authorization models, RBAC configurations, API access controls, or evaluating whether every request is authorized before access is granted.

LibAFL is a modular fuzzing library for building custom fuzzers. Use for advanced fuzzing needs, custom mutators, or non-standard fuzzing targets.

Coverage-guided fuzzer built into LLVM for C/C++ projects. Use for fuzzing C/C++ code that can be compiled with Clang.

Review and restrict Linux capabilities in Kubernetes containers. Use when writing, reviewing, or auditing pod security contexts that add or drop Linux capabilities such as BPF, NET_ADMIN, NET_RAW, SYS_ADMIN, IPC_LOCK, SYS_NICE, or SYS_PTRACE.

Support OAuth 2.0 Dynamic Client Registration in MCP clients. Use when building or reviewing MCP client registration with authorization servers that support dynamic registration.