learnerLj/agent-skills
Skill Claude CodeCodex
A Chinese-language workflow for locating and patching server-side feature gates in the Claude Code Bun binary, including workflow and ultracode access checks. A feature gate is a software switch that controls access to a feature.
learnerLj/agent-skills
Skill Claude CodeCodex
Create, tighten, or review runnable /goal contracts for coding agents. Turns vague engineering intent — issues, PRs, bug reports, plans, CI failures, backlog items — into one measurable task contract with hard constraints, auditable verification, and stop rules. Use when the user asks to write, generate, improve…
learnerLj/agent-skills
Skill Claude CodeCodex
Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.
learnerLj/agent-skills
Agent
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.
learnerLj/agent-skills
Agent
Compare two outputs WITHOUT knowing which skill produced them.
learnerLj/agent-skills
Agent
Evaluate expectations against an execution transcript and outputs.
learnerLj/agent-skills
Skill Claude CodeCodex
Run an extremely strict maintainability review for abstraction quality, giant files, and spaghetti-condition growth. Use for a thermo-nuclear code quality review, thermonuclear review, deep code quality audit, or especially harsh maintainability review.
learnerLj/agent-skills
Skill Claude CodeCodex
Verify a claim with fresh local evidence: restate it falsifiably, capture baseline and treatment, compare artifacts, and return VERIFIED, NOT VERIFIED, or INCONCLUSIVE.