AI Engineering

Yao Meta creates, evaluates, packages, and governs reusable agent skills from workflows, prompts, SOPs, and scripts, used for migration, release packaging, routing, evals/tests, and install/trust checks.

Agent Skills Platform turns descriptions, docs, scripts, and transcripts into tested, installable agent skills. Use it when you want to automate a workflow, package a custom skill, or validate and audit a skill before install.

Observal Agents creates, validates, publishes, versions, and manages Observal Agents from the command line, including pulling, archiving, restoring, and transferring ownership. Use it when building, installing, or changing an Agent definition or managing its lifecycle.

Observal Registry CLI submits, browses, installs, edits, archives, restores, transfers, and manages co-authors for portable skill packages. It helps users manage AI agent skills across harnesses like Claude Code and pi.

10x Mode finds 10x product opportunities and high-leverage improvements. Use it when you need strategic product thinking, want to find high-impact features, or ask what would make a product 10x better.

Agent MD Refactor restructures bloated agent instruction files like AGENTS.md and CLAUDE.md into concise root files with linked, categorized docs. Use it when an AI coding agent's instructions have grown too long, redundant, or contradictory.

Codex runs OpenAI Codex CLI for code analysis, refactoring, and automated editing, defaulting to GPT-5.2. Use it when a user asks to run codex exec or resume a Codex session.

Excalidraw delegates all operations on .excalidraw files to subagents so the main agent never reads verbose diagram JSON directly, preventing context exhaustion. Use it whenever a diagram needs to be read, created, modified, or compared.

Gepetto creates detailed, sectionized implementation plans through research, stakeholder interviews, and multi-LLM review. Use it when planning features that need thorough pre-implementation analysis.

Surveys the Carbon codebase as a senior advisor to produce prioritized, self-contained implementation plans for other agents to execute. Ideal for finding improvement opportunities and generating handoff plans without modifying code.

Automates the iterative development of a single, tightly-scoped work item (bug, tweak, small feature) by running a doer→gate→judge cycle unattended, producing a gated pull request as the only human supervision point.

Creates or refreshes an AGENTS.md for Carbon packages or ERP modules by reading source code, grounding every claim in real functions, tables, and exports. Ideal when adding modules or performing freshness audits.