Skill Claude CodeCodex
Summarize machine learning papers with claims, methods, metrics, and limits.
Verification-gated skill routing and self-improvement harness for Hermes-style agent skills
This repository also configures its own agents. See what hermes-skilleval tells them →
Skill Claude CodeCodex
Summarize machine learning papers with claims, methods, metrics, and limits.
Skill Claude CodeCodex
Rerank retrieved candidates with a pairwise neural relevance model.
Skill Claude CodeCodex
Adapt embedding models using labeled retrieval failures and contrastive pairs.
Skill Claude CodeCodex
Answer questions using retrieved context with source-grounded reasoning.
Skill Claude CodeCodex
Build and query embedding indexes for semantic retrieval.
Skill Claude CodeCodex
Use when reviewing local browser screenshots for layout shifts, visual regressions, and viewport state.
Skill Claude CodeCodex
Use when diagnosing failing tests, stack traces, and repeated errors with a hypothesis-driven debug loop.
Skill Claude CodeCodex
Use when writing failing tests first, running red-green cycles, and keeping regression tests focused.
Skill Claude CodeCodex
Use this for tasks and general help.
Skill Claude CodeCodex
Use when reviewing release notes for evidence boundaries, local validation summaries, and explicit non-goals.
Skill Claude CodeCodex
Use when auditing validation workflow evidence before a maintainer review or diagnostic handoff.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: