NeVibeCode · repository · 17 items
aegis: a set by ganyuanran
The ganyuanran/aegis repository on GitHub: 17 cards with a description and an install command. They work in Claude Code, Codex, Cursor — 8 platforms in total.
About the repository
- GitHub
- ganyuanran/aegis
- Stars
- ★ 1,450
- License
- MIT
- Source updated
- 2026-11-04
- In the catalog since
- 2026-10-26
writing-plans skill writing-plans transforms approved requirements into a step-by-step implementation plan with TDD routing before modifying repository files.
CX Codex +1
★ 1,450repo stars
verification-before-completion skill verification-before-completion is a Claude Code skill that checks current evidence before an agent reports work done, fixed, passing, or ready to ship. Used for tests and handoffs.
CX Codex +1
★ 1,450repo stars
using-git-worktrees skill using-git-worktrees — a Claude Code skill that decides whether a git worktree is actually needed before creating one. It checks repository state and only adds a checkout when a concrete conflict exists.
CX Codex +1
★ 1,450repo stars
using-aegis skill using-aegis is a Claude Code skill that routes Aegis requests to specialized workflows and keeps straightforward requests on the fast path.
CX Codex +1
★ 1,450repo stars
update-aegis skill update-aegis is a Claude Code skill that checks and updates the installed Aegis Method Pack, then verifies the result.
CX Codex +1
★ 1,450repo stars
ui-ux-governance skill ui-ux-governance — a skill that gives the agent rules for interface work: how to pick flows, handle states and keep accessibility intact.
CX Codex +1
★ 1,450repo stars
systematic-debugging skill The systematic-debugging skill guides bug and test-failure investigations before code changes. Use it to trace a symptom to the logic that owns the cause.
CX Codex +1
★ 1,450repo stars
recording-architecture-decisions skill The recording-architecture-decisions skill decides when an architecture decision belongs in an ADR and checks whether the current-state baseline still matches the change.
CX Codex +1
★ 1,450repo stars
long-task-continuation skill long-task-continuation maintains checkpoints, evidence records, and drift tracking across context resets or multi-agent workflows. Keeps state intact during multi-step coding tasks.
CX Codex +1
★ 1,450repo stars
goal-framing skill The goal-framing skill defines a task goal, evidence of success, and boundaries before work begins. Use it when the agent needs a clear stop condition and a route for the task.
CX Codex +1
★ 1,450repo stars
first-principles-review skill first-principles-review is a Claude Code skill that checks decisions against first principles and Occam's razor before an agent locks in a direction. It skips ordinary bug fixes.
CX Codex +1
★ 1,450repo stars
finishing-a-development-branch skill finishing-a-development-branch is a Claude Code skill that verifies tests and Git status when wrapping up a task, then merges, preserves, or deletes its branch and worktree; deletion requires exact confirmation.
CX Codex +1
★ 1,450repo stars
establishing-project-context skill establishing-project-context keeps domain terms in CONTEXT.md so people and agents call the same things by the same names. Useful when a term gets renamed, deprecated or overloaded.
CX Codex +1
★ 1,450repo stars
dispatching-parallel-agents skill dispatching-parallel-agents provides clear gating rules to run concurrent subagents on ad-hoc, decoupled tasks. Prevents file collisions and coordination waste when working without a written plan.
CX Codex +1
★ 1,450repo stars
communicating-concisely skill The communicating-concisely skill makes agent replies more compact when asked, while preserving technical details. The style stays active until the user turns it off.
CX Codex +1
★ 1,450repo stars
brainstorming skill brainstorming — a skill for Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, and OpenCode that blocks implementation until a design is approved. Triggers on ambiguous features, architecture decisions, and API contract changes.
CX Codex +1
★ 1,450repo stars
anti-entropy-governance skill anti-entropy-governance governs code retirement, eliminates duplicate logic, and stops automated deletion of database tables or persistent state without explicit confirmation.
CX Codex +1
★ 1,450repo stars