19 tagged self improving agents, measured the same way as everything else here.
Browse within: llm-agents 6skill-optimization 6flashinfer 5mlsys-2026 5nvidia-gpu 5
Skill Claude CodeCodex
Optimize a skill by planning, exploring available tools, diagnosing failures, and implementing fixes via PR. Use this skill as soon as an optimization loop starts, especially on the first iteration when community tools or reference approaches should be explored before implementation.
Skill Claude CodeCodex
Interactively gather a PRD (Product Requirements Document) for an agent skill that is about to be built or optimized by SkillHone. Use when the user says things like "I want to write a new skill", "help me spec this skill", "what should -skill do", or before running skillhone new / skillhone optim on a skill whose…
Skill Claude CodeCodex
SkillHone — toolkit for evaluating, optimizing, and managing agent skills. Use when asked to "evaluate a skill", "run probe", "optimize/iterate a skill", "create a new skill experiment", "seed a skill repo", or "run skill benchmarks"; also use when the user mentions a Forgejo-hosted skill repo and wants to measure or…
syhya/mlsys26-flashinfer-contest
Skill Claude CodeCodex
Optimize MLSYS26 FlashInfer contest operators for NVIDIA B200/Blackwell with a reference-first workflow. Use when tasks require refreshing only the current project's local reference/ git repos before optimization, searching high-quality official repos when local references are insufficient, deriving workload regimes…
syhya/mlsys26-flashinfer-contest
Skill Claude CodeCodex
Create and push FlashInfer contest submission tags such as submission-v3 in a git repository. Use when the user asks to "tag your submission", "create a submission tag", "push the latest submission tag", or prepare a FlashInfer contest repo so the evaluator will use the latest tagged commit. Always validate…
syhya/mlsys26-flashinfer-contest
Skill Claude CodeCodex
Code review and debugging assistant. Identifies bugs, performance issues, security vulnerabilities, and suggests optimizations.
Skill Claude CodeCodex
Onboards loopcraft into a project — scans the repo to detect gate commands, confirms the backlog and rubrics through an interview, then scaffolds the .loop/ structure. Run once per project, or when refreshing the config.
Skill Claude CodeCodex
Traverses the backlog unattended, applying the loop-task cycle (rubric · verifier · gate · Loop-Verified commit) to each item. Invoke only when the user explicitly starts an autonomous batch run — for individual work, use loop-task directly.
Skill Claude CodeCodex
Runs non-trivial implementation/fix work that needs a verifiable output through a rubric + independent verifier grading cycle. Invoke before starting code/docs work. Not for simple questions, exploration, or one-line fixes.