aptratcn

Pre-install security audit for third-party AI agent skills. Detects malicious patterns, social engineering, and permission overreach in SKILL.md files before installation.

Cognitive Debt Guard: Prevents a 23.5% incident spike from AI-generated code through comprehension gates, review frameworks, and AI-free zones, based on 2026 research.

Control your AI coding agent costs. Track, budget, and optimize your spending to avoid surprise bills.

Prompt Injection Guard detects and resists prompt injection in user input, web content, and AI outputs, ensuring security-first AI interactions.

The 4R error recovery framework (Recognize, Rescue, Report, Remember) ensures you never lose work silently. It triggers on terms like 'error', 'failed', 'crash', 'something went wrong', and 'retry'.

See where your tokens go. Retroactive analysis, hotspot detection, and cost optimization for AI agents. Zero dependencies.

Interactive codebase health audit covering 8 dimensions: Architecture, Security, Performance, Maintainability, Testing, Docs, Dependencies, Quality. Zero dependencies.

Security audit for MCP (Model Context Protocol) servers. Detects data exfiltration risks, command injection, permission escalation, and supply chain vulnerabilities before adding MCP servers to your agent.

Automatically generates a structured project knowledge base (CONTEXT.md) from codebase analysis, creating a living map of architecture, conventions, decisions, and dependencies. It has zero dependencies, uses pure markdown, and is inspired by mattpocock's shared language and jcode's semantic memory, but automated and structured.

Give your AI agent a soul, defining personality, values, taste, and growth in pure markdown. Create soul-driven agents with zero dependencies.

Test prompts for regression after changes. Catch quality drops before they hit production. Zero dependencies.

Lightweight session checkpoints for AI coding agents. Save critical context at key moments, restore when interrupted. Zero dependencies.

Debug, test, and troubleshoot MCP servers. Verify tool schemas, test tool calls, and diagnose connection failures. Zero dependencies.

Code migration and upgrade assistant with 7 technology playbooks, breaking change detection, and rollback planning. Zero dependencies.

Give your AI agent a consistent brand voice through 8-stage brand therapy, tone calibration, and content templates, with zero dependencies.

GitOps Workflow Patterns for AI Agents. Reliable deployment, rollback, and infrastructure management. Blue-green, canary, CI/CD templates, health checks — pure markdown, zero dependencies.

Token Budget Guard: Stop burning context. Achieve 99% token savings through progressive disclosure, conversation compression, and budget monitoring, ensuring native token awareness for every agent workflow.