Use when a training run needs evidence about convergence, objective, gradients, regularization, search budget, batch or precision strategy.
Plugin yzmir-training-optimization from marketplace foundryside-marketplace in tachyon-beep/skillpacks. Commands read from the repo's manifest — 1 month ago.
using-training-optimizationUse when a training run needs evidence about convergence, objective, gradients, regularization, search budget, batch or precision strategy.
/yzmir-training-optimization:check-gradientsCheck gradient flow and update health with task-specific source and runtime evidence.
/yzmir-training-optimization:diagnoseDiagnose a training run with task-specific source and runtime evidence.
/yzmir-training-optimization:setupConfigure a bounded training baseline with task-specific source and runtime evidence.
training-config-reviewerReview training configuration in context with task-specific source and runtime evidence.
training-diagnosticianDiagnose training behavior with task-specific source and runtime evidence.
Use when drafting, revising, critiquing, or planning prose fiction or creative nonfiction, especially when author voice, continuity, genre expectations, or truth claims need deliberate treatment.
1 skill · 3 commands · 11 agents
Use when documentation accuracy, executable procedures, institutional register, sensitive information, or source-to-document traceability needs deliberate review.
2 skills · 4 commands · 5 agents
Use when reconstructing an existing codebase: source-linked architecture, dependency and risk findings, coverage limits and incremental refresh.
1 skill · 5 commands · 4 agents
Use for Rust-specific ownership/API, async cancellation, unsafe/FFI soundness, feature/toolchain compatibility, diagnostics, packaging or profiling problems in a real crate.
1 skill · 5 commands · 3 agents
Use when an organization explicitly selects a software governance profile requiring lifecycle evidence, traceability, risk decisions or measurement.
5 skills · 4 agents
Use when test reliability, isolation, coverage gaps, performance experiments, resilience, or release evidence needs investigation or a concrete verification strategy.
1 skill · 5 commands · 3 agents
Official Next.js skills: adopt and optimize Cache Components, adopt Partial Prefetching, and verify runtime behavior against a running dev server.
6 skills
Token usage tracking, model cost attribution per agent, budget alerts, and optimization recommendations
24 skills · 1 command · 1 agent · hooks
Transform Claude into an Azure expert. This plugin integrates the Azure MCP server and specialized Azure skills to move beyond generic advice. It enables Claude to perform real-world tasks: listing resources, validating deployments, diagnosing infrastructure issues, and optimizing costs across 50+ Azure services.
44 skills · 1 MCP server
Eval-gated LLM fine-tuning lifecycle (LoRA/QLoRA, DPO, GRPO/RLVR, vision SFT, quantized export). Use when fine-tuning an open-weights model end-to-end: building the eval harness first, selecting method and model, preparing data, training with Unsloth or TRL, gating checkpoints, and exporting GGUF/FP8.
10 skills · 2 commands · 3 agents
CockroachDB plugin for Claude Code — explore schemas, write optimized SQL, debug queries, and manage distributed database clusters directly from your AI coding agent.
34 skills · 3 agents · 3 MCP servers · hooks
Per-deal-and-packaging Commercial domain. v2.8.0 ships 8 skills: orchestrator + pricing-strategist (model picker + Van Westendorp WTP + packaging) + deal-desk (deal scorer + discount approval routing + redline) + partnerships-architect (5-tier classifier + joint GTM + revshare modeler) + channel-economics (cost-to-serve + ROI + mix optimizer) + commercial-policy (data-backed discount matrix + exception flow + linter) + rfp-responder (Shipley structured RFP/RFI/RFQ + win-theme + winrate predictor, context: fork) + commercial-forecaster (4Q-weighted bookings + cohort NRR/GRR + funnel-confidence with mandatory assumption disclosure). Hard rules: pricing outputs model+range (never a single number), deal outputs route to named human approver (never auto-approve), forecast outputs surface conversion assumption, RFP never invents claims for GAP requirements. 21 stdlib Python tools, 28+ reference docs. Distinct from business-growth (sales execution), c-level-advisor/cro-advisor (strategic), finance (close+report).
8 skills · 9 commands · 1 agent
This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.