Skill Building

Scaffold eval specs with stimuli, graders and rubrics for agent skills and workflows

Diagnose and fix agent skills that fail to activate, time out or lose to baseline

Create, audit and tidy up AgentSkills and their SKILL.md files

Search, install and upgrade SkillHub skills with the official SkillHub CLI

Create or update agent skills with clear scope, workflows and bundled resources

Score SKILL.md files against specs and best practices with fix suggestions

Find out which skill or workflow fits your situation in a set of engineering skills

Create, evaluate, package and upload agent skills with eval runs

Search the open skills.sh ecosystem and install agent skills you need

Create, improve and benchmark agent skills with evals and description tuning

Migrate skills from Claude Code, OpenClaw or Hermes folders into PilotDeck

Send feedback, bug reports and feature requests about the TRES plugin

Extract reusable skills and lessons from a finished work session as new skills

Map OKR-style agent workflows and skill discovery for x-cmd

Write skill documents with pyramid structure, line limits and layout rules

Author a SKILL.md for OpenKnowledge or your editor, then tune its triggering

Turn the current session into a reusable skill saved as a candidate

Build and publish Kiro Crew apps with manifests, UI components, crons and hooks

Find and reuse org-approved skills, prompts, MCP servers and agents for a task

Create, publish, version and transfer Observal agent definitions and ownership

Write agent skills that behave the same way on every run

Install or remove third-party skills from a git repo with a safety gate

Build, install and debug bb plugins for panels, CLI commands and themes

Sync global skills across Codex, Claude Code, Pi, Hermes and Cursor

Write, review and debug SKILL.md files and their invocation problems

Push skills and AGENTS.md to a private repo and verify the public mirror

Upgrade the NotFair plugin to its latest version and refresh its cache

Recommend which agent skills a project needs based on its past sessions

Design and write production-grade skills with solid workflows and packaging

Audit a skill for structure, description quality, token use and anti-patterns

Record a Linux desktop or browser workflow and turn it into a reusable skill

Find the right NVIDIA skill for a GPU, Jetson, CUDA, NIM or Isaac request

Generate a governance skill card for an existing agent skill folder

Route any Go coding task to the most relevant Go skills automatically

Turn sketch design findings into a persistent project skill

Turn spike findings into a persistent project skill for later builds

Load a bundle of 127 skills with a dashboard, burn report and session diary

Scaffold Claude plugins, skills and agents with the right structure

Create new agent skills and share them with your team on Slack

Write and test GoClaw agent skills with evals, benchmarks and description tuning

Turn mature curated wiki pages into a reviewable agent skill package

Log corrections and patterns during tasks and turn them into skill updates

Turn your editing style into a reusable skill file for future video edits

Review finished work and distill lessons into reusable methods, skills and evals

Write and check SKILL.md files and MCP tool descriptions for quality

Author and tighten SKILL.md skills with YAML frontmatter and install layouts

Create, improve and benchmark agent skills with evals and description tuning

Report repeated failed edits in fa-* skills back to the framework as feedback

Score skills with deterministic checks plus LLM review and gate merges on a threshold