codex

Cascade Evals

Designs and runs versioned evaluations for prompts, agents, simulations, and coding-agent harnesses with deterministic eligibility, calibrated independent judges, conservative reduction, and immutable evidence receipts.

2 on GitHubby Cascadev0.3.29+codex.20261005203443updated 6 days ago

Ships a Codex manifest but no marketplace file, so Codex can't add it by name yet. Add it to a local marketplace (.agents/plugins/marketplace.json).

Plugin cascade-evals in oleksandr-rud/cascade-codex. Commands read from the repo's manifest — 6 days ago.

What it installs

Skills (6)

  • agent-evaluation

    Adapt the generic Cascade Evals lifecycle to an AI agent, role, skill, workflow, tool loop, or architecture packet. Use for agent-task corpora, trace eligibility, outcome and trajectory judging, version comparison, or improvement evidence; keep agent-architecture-specific packs with their owning Cas

  • build-judge

    Create, revise, version, validate, or calibrate semantic judge profiles, anchored rubrics, response schemas, labeled cases, and aggregation rules for prompt, agent, simulation, or harness evaluations. Use when semantic quality cannot be decided mechanically; do not use to execute targets or judge a

  • evaluate

    Design, validate, reduce, or audit a versioned evaluation across prompts, agents, simulations, or coding-agent harnesses. Use when a subject needs deterministic eligibility, independent semantic judges, conservative aggregation, calibrated evidence, or an immutable evaluation receipt; use a subject

  • harness-evaluation

    Adapt Cascade Evals to coding-agent harness skills, routes, roles, outputs, and execution traces. Use for bounded trigger, collision, eligibility, outcome, trajectory, or regression diagnostics; keep each target repository's scenario catalog and harness contracts with that repository.

  • prompt-evaluation

    Run controlled prompt and adaptive-interview evaluations through bounded model execution, and design Laya/Jev/Intern-Decision/Imajev typed-decision evaluations. Use when prompt quality, typed question accuracy, model-tier fit, repeated variance, interview behavior, independent judgment, or human cal

  • simulation-evaluation

    Independently evaluate the semantic outcome, policy support, or claim support of one frozen Cascade Simulations run after controller verification. Use when a verified actor run needs blind judging or conservative evidence reduction; do not execute, repair, or redefine the simulation.

View source on GitHub

More from oleksandr-rud/cascade-codex

All 16 plugins →

Cascade AI Architect

codex

Designs source-grounded AI agents and agentic systems while delegating prompt authoring, evaluation execution, simulation, and harness integration to their owning plugins.

11 skills

2

Cascade Simulations

codex

Compile actors, manage versioned campaigns, execute bounded goal-directed simulations through Codex-host tools, and review frozen runs against fixed contracts.

9 skills

2

Cascade Coding Agent

codex

Integrates develop or the primary branch, resolves conflicts by authority, prepares reviewed changes for push, and engineers coding-agent harnesses.

5 skills

2

Cascade Design

codex

Apply proportional design to UI changes and refactors: appearance, components, forms, parameter configuration, search, navigation, layouts, workflows, responsive behavior and data/canvas interfaces; create mockups and review visual/accessibility evidence.

5 skills

2

Cascade Project Management

codex

Define tracker-ready work items and plan or coordinate evidence-bound projects and lean Agile MVP delivery without taking product priority, filing, or target execution authority.

4 skills

2

Cascade QA

codex

Plan quality, design tests, assess evidence, and triage defects from accepted behavior without taking product definition, host execution, test repair, or release authority.

4 skills

2

Similar plugins

career-ops

claudecodex

Open-source AI job search that runs locally in your AI coding CLI: scan job portals, evaluate listings into a structured A-H report with a 1-5 score, tailor your CV, track applications

5 skills · hooks

career-ops-hq/career-ops 82,253

academic-research-skills

claude

5 skills + 35 modes + Material Passport pipeline. Includes v3.6.7 cross-model audit gate and v3.6.8 generator-evaluator contract.

5 skills · 16 commands · 4 agents · hooks

Imbad0202/academic-research-skills 56,698

huggingface-skills

claude

Build, train, evaluate, and use open source AI models, datasets, and spaces.

26 skills · 1 MCP server

anthropics/claude-plugins-official 47,076

ifixai

claudecodex

Independent auditing for AI agents, open source. Claude discovers your setup, confirms the cost, and audits your agent against the one question other tools skip: is it doing the job it's supposed to do, given your business rules and org structure? Test any model (Anthropic, OpenAI, Gemini, Azure, Bedrock, …) graded by self, one independent judge, or a cross-vendor panel; each provider is billed to its own account via a key in your settings, never the command line. Results render as an interactive artifact (with a static-report fallback).

1 skill · hooks

ifixai-ai/iFixAi 45,158

LSEG

claude

Price bonds, analyze yield curves, evaluate FX carry trades, value options, and build macro dashboards using LSEG financial data and analytics.

8 skills · 8 commands · 1 MCP server

anthropics/financial-services 41,259

Maintain Cascade Evals?

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.

claude & codex plugins16 plugins