Skill UI

This skill reads the contents of a UTF-8 text file and outputs the raw bytes verbatim to stdout. It is designed for meta-skills that require round-tripping an artifact through disk for user inspection or manual editing between steps. Unlike standard file readers that prepend line numbers, this tool preserves the original structure, making it ideal for structured data formats like YAML, scripts, or SRT files that need to be accurately parsed downstream.

This workflow systematically analyzes deployed AWS resources and Infrastructure-as-Code (IaC) definitions. It collects usage metrics from CloudWatch and Cost Explorer to identify over-provisioning, unused resources, and inefficient configurations (e.g., right-sizing EC2, S3 lifecycle policies). It generates actionable cost optimization recommendations and creates structured GitHub issues for tracking and implementing savings.

This automated workflow performs a comprehensive review against the AWS Well-Architected Framework (WAF). It scans IaC definitions (Terraform, CloudFormation, CDK) and deployed resources to identify compliance gaps, security risks, and optimization opportunities across six core pillars. Findings are structured and automatically generated as actionable GitHub issues with remediation code examples.

This guide details the principles of observability, teaching developers how to instrument code by implementing structured logging, metrics, and distributed tracing. It ensures that production features emit sufficient telemetry to answer complex operational questions, transforming unknown failures into diagnosable events. Essential for any service running in production.

This skill provides expert, developer-level guidance on the Power BI Project (PBIP) file format. PBIP deconstructs the opaque PBIX binary into human-readable, structured text files, enabling robust source control, external editing, and multi-author collaboration. It covers project structure, the difference between thick and thin reports, file conversion (PBIX to PBIP), and critical cross-cutting concerns like rename cascades and data location discrepancies across various report components.

This advanced tool surfaces critical, often unrecorded, evidence of human-AI collaboration during development sessions. It analyzes git diffs and conversational context across multiple layers to classify the signal level (HIGH/MEDIUM/LOW), measure technical complexity, detect decision forks, and quantify the specific contribution of the AI model, providing structured data for post-session retrospectives and accountability.

This skill allows users to mine their extensive coding history across various AI agents (Claude, Codex, Gemini, etc.) into a persistent, structured corpus. It surfaces reproducible workflows as evidence and interactively helps the user forge these past practices into reusable, installed skills across their agent ecosystem. It is designed to synthesize judgment from raw transcripts into actionable, portable skills.

This Main Agent acts as a central orchestrator for a specialized squad of AI agents, managing complex software development lifecycles. It guides projects through defined phases—from requirements gathering and planning to architecture, implementation, testing, and deployment. It maintains strict context discipline by summarizing agent outputs and referencing full artifacts, ensuring efficient, structured, and non-overwhelming communication to the user.

Atlas Ledger acts as a crucial memory companion to atlas-contract. It automatically triggers when a deviation ('drift') is caught during audits or reviews. Its core function is to distill the observed error or mistake into structured, actionable IF/WHEN/DON'T/INSTEAD clauses. These clauses are written to a permanent ledger (Atlas.md), ensuring that subsequent interactions are governed by learned lessons, thereby strengthening the project's overall governance and reliability.