Skill Claude CodeCodex
Optimize an AI agent's harness for MCP-Atlas benchmark. Use when analyzing execution traces, diagnosing failures, and proposing improved prompts, skills, or harness code.
80 tagged self improving, measured the same way as everything else here.
Browse within: ai-development 36methodology 36afk 18harness-engineering 18continual-learning 16llm-agents 16recursive-self-improvement 16self-evolving 16
Skill Claude CodeCodex
Optimize an AI agent's harness for MCP-Atlas benchmark. Use when analyzing execution traces, diagnosing failures, and proposing improved prompts, skills, or harness code.
Skill Claude CodeCodex
When to prefer GUI mouse clicks over keyboard shortcuts — especially for formatting, multi-step visual tasks, and cross-application workflows.
Skill Claude CodeCodex
Reading, writing, and converting common data formats (CSV, Excel, JSON, YAML) with correct handling of encoding, types, and edge cases.
Skill Claude CodeCodex
Wire an AI agent's telemetry into Kyoko and prove one real run shows up.
Skill Claude CodeCodex
Before starting any non-trivial implementation, run a pre-flight reconnaissance pass to triangulate git state, project infrastructure, and prior-session memory. Auto-assembles a verified grounding preamble — a session-scoped artifact the orchestrator pastes verbatim into every subsequent sub-agent brief — eliminating…
Skill Claude CodeCodex
Dispatch a parallel adversarial verifier wave after any high-stakes sub-agent investigation (code reviews, audits, findings reports, large refactors, gap analyses) — or whenever a sub-agent asserts a claim with high-confidence language ("confident", "certain", "clearly", ≥80%), since confidence is a trigger, not a…
Skill Claude CodeCodex
Release pipeline for already-done local work. Dispatches /ground-state pre-flight, runs the project test suite, drafts a commit message, pushes, and opens a PR with a structured verification summary. Use when local changes are ready to hand off to review — e.g. 'ship this', 'push and open a PR', 'release this work'.…
Skill Claude CodeCodex
Create a new reusable agent skill — either explicitly ("turn this workflow into a skill") or by reflecting on a just-completed task that you might do again. Synthesizes SKILL.md, scripts, and a real eval scaffold, then stages it for review before committing. Use when the user asks to make/save/capture a skill, or…
Skill Claude CodeCodex
Find the right agent skill for a task from public sources (curated index, GitHub repos, and locally installed skills), then install the chosen one. Use when the user asks to find, search for, or discover a skill, or when a task would benefit from a skill you don't have yet.
Skill Claude CodeCodex
Improve an existing skill through an evaluation-gated loop — run it on its eval tasks, analyze failures, propose bounded edits, validate a candidate, and (with your approval) promote a strictly-better version. Use when a skill underperforms or the user asks to optimize/tune/improve a skill that has an eval set.
Skill Claude CodeCodex
Multi-agent orchestration framework for high-intensity feature implementation. Routes work across model tiers: frontier models (Claude Fable, Kimi 3, Opus; Codex only when explicitly selected) own design, architecture, and review sign-off, while the pinned Luna Max economy lane handles implementation and mechanical…
stepanenkoviktor0110-boop/ai-dev-methodology
Skill Claude CodeCodex
Universal quality coding process: plan, TDD, reviews. Use whenever code needs to be written — ad-hoc or as part of a task. Use when: "напиши код", "закодь", "реализуй", "write code", "implement" For planning tasks → tech-spec-planning skill. For specs → user-spec-planning skill.
stepanenkoviktor0110-boop/ai-dev-methodology
Skill Claude CodeCodex
A feature-delivery skill that coordinates coding agents in stages, handles review cycles, and creates a commit for each stage.
stepanenkoviktor0110-boop/ai-dev-methodology
Skill Claude CodeCodex
A skill for resolving conflicting development requirements, where one part of a system is expected to satisfy opposing conditions. It looks for existing resources in the code and data structure, then tests possible ways to separate the conflict by structure, timing, conditions, or relationships.