Agent Skills by Feng Jiang
Showing 30 vetted skills indexed across 1 GitHub repositories.
GSU-FrankJusing-git-worktrees
Create isolated Git worktrees with directory safety checks and automated setup.
GSU-FrankJmulti-agent-patterns
Design multi-agent systems with supervisor, swarm, and hierarchical patterns.
GSU-FrankJbdi-mental-states
Model agent mental states by converting RDF knowledge graphs into BDI beliefs, desires, and intentions.
GSU-FrankJcontext-degradation
Diagnose AI context degradation and provide mitigation strategies.
GSU-FrankJtest-driven-development
Enforce Test-Driven Development by mandating failing tests before production code.
GSU-FrankJsystematic-debugging
Guide software debugging through root cause investigation, pattern analysis, hypothesis formation, and implementation.
GSU-FrankJcontext-compression
Compress AI conversation history and codebases to reduce token counts.
GSU-FrankJmemory-systems
Implements agent memory systems for cross-session persistence and structured knowledge representation via temporal graphs, vector stores, and entity tracking.
GSU-FrankJadvanced-evaluation
Implement LLM-as-a-judge evaluation pipelines with bias mitigation and pairwise comparison.
GSU-FrankJusing-superpowers
Enforces skill invocation before any AI response or action.
GSU-FrankJdispatching-parallel-agents
Dispatch independent agents to concurrently debug multiple unrelated failures.
GSU-FrankJexecuting-plans
placeholder
GSU-FrankJfinishing-a-development-branch
Verify tests, then merge branches, create pull requests, or clean worktrees.
GSU-FrankJcontext-fundamentals
Explain context engineering concepts for AI agent systems.
GSU-FrankJbrainstorming
Transform user ideas into detailed project designs and specifications.
GSU-FrankJproject-development
Guide LLM project planning from task assessment to pipeline architecture and cost estimation.
GSU-FrankJevaluation
Automate AI agent evaluation with multi-dimensional rubrics and LLM-as-judge.
GSU-FrankJwriting-plans
Generate step-by-step implementation plans with file paths, code snippets, and commit messages.
GSU-FrankJrequesting-code-review
Automate code review requests by identifying Git changes and dispatching a subagent.
GSU-FrankJreceiving-code-review
Guide AI agents to verify and respond to code review feedback.
GSU-FrankJcontext-optimization
Summarize conversation turns and mask tool outputs to reduce token usage.
GSU-FrankJwriting-skills
Create and verify AI Skills using a Test-Driven Development methodology.
GSU-FrankJrunning-experiments
Execute and analyze game-theoretic experiments using PPO and gradient methods.
GSU-FrankJverification-before-completion
Enforce command verification before declaring tasks complete or builds successful.