executing-plans
executing-plans
Use when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available
Skills entdecken
Produktivität · SKILL.md
rl-verifiable-rewards
rl-verifiable-rewardsRLVR extends reasoning capabilities by proving answer-only rewards implicitly incentivize correct intermediate reasoning via the Logic Prior principle.
Independent GitHub snapshot · Automated classification · Original documentation governs
Skill-Dokumentation
GitHubDokumentation wird aus der Quelle geladen…