Agent Skills by ginmp
Showing 32 vetted skills indexed across 1 GitHub repositories.
ginmp8skill-harness
Audit, validate, and harden ChatGPT or agent skills across multiple modes.
ginmp8skill-package-architecture-review
Evaluate skill package architecture against SKILL.md and resource criteria.
ginmp8frontend-ai-token-efficient
Provide contract-driven guidance for React/TypeScript frontend architecture reviews and validation.
ginmp8skill-hardening
Audit and tighten skill package frontmatter, resources, validators, outputs, and packaging.
ginmp8nuget-package-updater
Validate NuGet package updates in Directory.Packages.props and generate a Markdown decision document.
ginmp8skill-hypothesis-discovery
Prioritize testable improvement hypotheses from skill evidence into a backlog.
ginmp8streamlit
Build, test, package, and deploy Streamlit apps with templates and helper scripts.
ginmp8skill-testing-and-validation
Validate skill package integrity with baseline evidence and structured failure classification.
ginmp8skill-token-efficient
Compress skill instructions to reduce token cost while preserving semantics.
ginmp8documentation-quality
Review documentation against SKILL.md contracts and reference guidelines.
ginmp8modern-dotnet-engineering
Provide production-grade .NET 10 engineering guidance for architecture, code review, and delivery.
ginmp8skill-improver
Orchestrate hypothesis-driven skill package improvements with frozen evaluators and rollback.
ginmp8skill-benchmark
Generate standardized benchmark reports for reusable agent skill packages.
ginmp8skill-consistency-repair
Audits and repairs Skill packages for consistent activation, ownership, workflow, resources, validation and packaging via inventory_skill.py and consistency_audit.py.
ginmp8skill-change-gate
Classify regressions in proposed skill package changes and return a pass, fail, or warning decision.
ginmp8sequential-work-packaging
Plan and package sequential work into canonical artifact structures.
ginmp8skill-creator
Create or update modular Skill packages with SKILL.md frontmatter and resources.
ginmp8agent-design
Design and validate custom agents, prompts, and governance structures for software projects.
ginmp8skill-prompt-and-activation-review
Review and rewrite activation descriptions, boundaries, and output contracts for prompts.
ginmp8thinking-token-efficient
Orchestrate compact private reasoning with citations and validation artifacts.
ginmp8magia
Automate bounded execution and evidence-driven engineering across code, tests, and documentation.
ginmp8mago
Generate canonical planning artifacts under BOARD_ROOT/specs/<spec_id>/ from governance intake and repository truth.
ginmp8karpathy-guidelines
Provide bounded, verifiable guidance for code review, debugging, refactoring, and testing.
ginmp8context-architect
Map repository context and dependencies before multi-file edits.