By makash
Respond to supply chain security incidents across npm, PyPI, GitHub Actions, AWS, GCP, Azure, and multi-language projects by generating interactive triage checklists, incident runbooks, shell scripts for credential rotation, IOC detection in logs, exposure audits, remediation, and dependency hardening with pinning, SBOM, and signing.
Detect whether stolen credentials were used and rotate them after a supply chain attack or security incident. Use this skill when credentials may have been exfiltrated and the user needs to determine if they were abused, rotate compromised credentials, or verify rotation completeness. Trigger when users ask about checking cloud audit logs after a compromise, detecting unauthorized credential use, finding lateral movement from stolen tokens, rotating credentials after an incident, auditing API key usage, or verifying that credential rotation was complete. Also trigger when an ecosystem-specific skill (pypi-supply-chain-response, npm-supply-chain-response, github-actions-supply-chain-response) hands off credential rotation to this skill. Works as a follow-up to any incident response skill or standalone for credential-focused incidents.
Respond to compromised GitHub Actions where tags have been overwritten with malicious code. Use this skill when a GitHub Action is reported compromised, when CI/CD secrets may have been exfiltrated through a poisoned action, when someone mentions Trivy, Checkmarx KICS, or any GitHub Action being backdoored, hacked, or tampered with. Also trigger when users ask about checking workflow run logs for exfiltration indicators, rotating CI secrets after a GitHub Actions compromise, or auditing GitHub Actions references across their organization.
Respond to npm supply chain attacks and compromised package incidents. Use this skill whenever a user mentions a compromised npm package, an npm supply chain attack, a malicious dependency in node_modules, credential-stealing malware from npm install, or asks how to check if they're affected by a package compromise on npm. Also trigger when the user asks about postinstall script backdoors, typosquatted npm packages, hunting for IOCs after an npm install, auditing node environments for malicious packages, or generating an incident response checklist for an npm compromise. Trigger even if the user just names a package and says it was "hacked", "backdoored", "compromised", or "pwned" and the package is from npm, yarn, or pnpm. Covers axios, plain-crypto-js, and any future npm supply chain incident.
Respond to Python/PyPI supply chain attacks and compromised package incidents. Use this skill whenever a user mentions a compromised Python package, a PyPI supply chain attack, a malicious dependency, credential-stealing malware in a pip package, or asks how to check if they're affected by a package compromise. Also trigger when the user asks about rotating credentials after a Python package incident, finding transitive dependencies, hunting for IOCs from a pip install, auditing Python environments for malicious packages, or generating an incident response checklist for a PyPI compromise. Trigger even if the user just names a package and says it was "hacked", "backdoored", "compromised", or "pwned".
Proactively audit and harden dependency management against supply chain attacks. Use this skill when a user asks about securing their dependencies, hardening their CI/CD pipeline against supply chain attacks, auditing their lockfiles or dependency pins, setting up SBOM generation, implementing dependency signing or provenance verification, or preventing the next supply chain compromise. Also trigger proactively when reviewing dependency configuration files (package.json, requirements.txt, Gemfile, go.mod, Cargo.toml, pom.xml) and noticing risky patterns like unpinned versions, missing lockfiles, or postinstall scripts. This skill is preventive — for active incident response, use the ecosystem-specific skills instead.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Security skills for AI coding agents. When your dependencies get compromised, these skills are the incident response playbook your agent follows.

March 19, 2026 — Trivy. Attackers compromise 76 of 77 tags on aquasecurity/trivy-action. Every GitHub Actions workflow using a tag reference runs attacker-controlled code. CI secrets — cloud credentials, deploy keys, package registry tokens — are exfiltrated via dead-drop repos.
March 23, 2026 — KICS. Using credentials stolen from the Trivy attack, attackers pivot to Checkmarx KICS, overwriting 35 tags on checkmarx/kics-github-action. The cascade continues.
March 24, 2026 — LiteLLM. Credentials stolen from the KICS compromise are used to publish backdoored versions of LiteLLM on PyPI (1.82.7, 1.82.8). The malware drops a .pth file in site-packages/ — Python executes it on every interpreter startup, before your code even imports. SSH keys, AWS credentials, .env files, everything is swept and exfiltrated. Most affected developers never directly installed LiteLLM — it was pulled in transitively by CrewAI, DSPy, and Browser-Use.
March 31, 2026 — Axios. The npm maintainer account jasonsaayman is compromised. Malicious versions [email protected] and [email protected] are published, injecting a typosquatted dependency plain-crypto-js that deploys platform-specific backdoors: a disguised binary on macOS (/Library/Caches/com.apple.act.mond), a renamed PowerShell on Windows (wt.exe), a Python script on Linux (/tmp/ld.py). The payload self-deletes its installer and swaps package.json to cover its tracks. Axios has 80 million weekly downloads.
One compromised account cascaded across three ecosystems in ten days. GitHub Actions → PyPI → npm. Each attack used credentials stolen from the previous one.
AI coding agents run pip install, npm install, and GitHub Actions workflows on your behalf. They pull dependencies, build containers, and deploy code. When a supply chain attack hits, the agent that installed the compromised package is also the fastest path to triage:
But agents don't know incident response by default. These skills teach them.
Six skills organized in layers:
npm-supply-chain-response — Deep triage for compromised npm packages. Built around the Axios attack: typosquatted dependency injection, multi-platform backdoors, anti-forensics (self-deleting setup.js, package.md swap). Six-phase workflow with three output modes (interactive checklist, runbook, automated script).
pypi-supply-chain-response — Deep triage for compromised PyPI packages. Built around the LiteLLM attack: .pth startup hooks, transitive dependency exposure via pipdeptree -r, credential harvesting. Cross-platform manual playbook (Windows PowerShell, macOS, Linux).
github-actions-supply-chain-response — Incident response for tag overwriting attacks. Built around the Trivy → KICS cascade: org-wide workflow scanning, run window confirmation, IOC hunting in CI logs, dead-drop repo detection.
credential-exfiltration-response — Full detection-to-rotation lifecycle. All ecosystem skills hand off here. Covers 13 credential classes (SSH, AWS with STS session invalidation, GCP, Azure, GitHub, npm, PyPI, Docker, Kubernetes, databases, .env secrets, CI/CD secrets, crypto wallets). Includes audit trail queries for every major cloud provider.
supply-chain-security-check — Multi-ecosystem fallback for Go, Rust, Ruby, Java, .NET, Docker, and incidents that span multiple ecosystems. Routes to ecosystem-specific skills when available.
supply-chain-best-practices — Nine-category dependency audit: version pinning, lockfile integrity, install hooks, vulnerability scanning, provenance verification, CI secret scoping, SBOM generation, update strategy, package manager hardening. Produces a PASS/WARN/FAIL checklist. Use this before an incident, not during one.
Add the marketplace in Claude Code:
/plugin marketplace add makash/agent-infra-security
Install the plugin:
/plugin install agent-infra-security@agent-infra-security
Reload plugins:
/reload-plugins
If you're outside Claude Code, prefix with claude:
npx claudepluginhub makash/agent-infra-security --plugin agent-infra-securityWrite high-converting website copy using proven frameworks (PAS, AIDA, BAB, StoryBrand). Auto-activates for landing pages, hero sections, CTAs, and product copy. Enforces outcome-driven headlines, 3:1 you-to-we ratio, banned word filtering, and conversion-optimized section flow. Use when writing website copy, landing pages, README intros, marketing pages, or any copy meant to convert visitors into users.
Audit and harden your software supply chain - packages, containers, GitHub Actions, IaC, AI/ML models, and IDE extensions. Action commands fix issues directly; walkthrough commands guide you through advanced setup.
Audit supply-chain threat landscape of project dependencies for exploitation or takeover risk
Comprehensive vulnerability scanning for code, dependencies, and configurations with CVE detection
Open-source cybersecurity analysis agent. Scans any local project for vulnerabilities: code security (SAST), dependency CVEs (SCA), secret leaks, authentication/authorization flaws, cryptographic weaknesses, misconfigurations, supply chain risks, and CI/CD security. Covers all OWASP 2025 Top 10 and CWE Top 25 categories. Generates prioritized reports with remediation guidance. Invoke with /cyber-neo [path].
AI-powered security auditing with interactive skills, automated agents, web dependency scanning, and supply chain hardening for comprehensive vulnerability detection and reporting
Offline security scanner for AI-agent repos, skills, plugins, and MCP servers