MARKET RESEARCH

CodexPluginStats: Codex plugin directory

Codex Plugin Stats: discover plugins, explore their skills and track catalog growth and observed changes.

Explore 5,323 plugins across 14 categories.

All plugins 7

1–7 of 7

All query words must match. Use quotes for an exact phrase. Matches include publisher text, keywords and attributed research.

Discover Codex plugins by task and category
Plugin / developer	Category	Follow
RUMBO Agent ReliabilityRUMBO IAEvidence-first workflows for reliable coding-agent execution, continuity, scope control, and final-state verification. Plugin name RUMBO Agent Reliability Show in context → 5 more matching sources Package name rumbo-coding-agent-reliability Show in context → Publisher keywords · listing coding agents agent reliability governance verification audit continuity Show in context → Publisher subtitle Evidence-first agent workflows Show in context → Publisher description Evidence-first workflows for reliable coding-agent execution, continuity, scope control, and final-state verification. Show in context → Publisher full description Recover authoritative state, reconcile conflicting evidence, control agent loops, verify real effects, and audit final state before accepting coding-agent work. Show in context →	ProductivityVersion 0.1.6	View details →
CRIADOR / ReefRONALD FERRARI SOARES Publisher full description Guides structured evaluation, reflection, and iteration for improving agent behavior, prompts, workflows, and reliability. Show in context → 6 more matching sources Package name reef-agent-improvement Show in context → Publisher subtitle Agent improvement workflows Show in context → Publisher capabilities · listing Agent evaluation Self-improvement workflow design Show in context → Publisher keywords · listing agents continual-learning self-improvement reef harness evaluation Show in context → Publisher name · package Human-Agent-Society Show in context → Publisher description Guidance and references for building continual self-improvement workflows for agents with Reef. Show in context →	Developer ToolsVersion 1.0.0	View details →
Frontier InfraTitanium ComputingHelp developers build standalone governed agent systems with Frontier Infra patterns and SDK components. Publisher keywords · listing agent-harness governance verification attestation agent-readable-web reliability Show in context → 2 more matching sources Publisher description Help developers build standalone governed agent systems with Frontier Infra patterns and SDK components. Show in context → Publisher full description Developer tooling that helps Codex build standalone governed agent applications without tying the target architecture to one model, coding harness, or dashboard. The coding assistant remains outside the runtime while it implements contracts, deterministic contr… Show in context →	ProductivityVersion 0.3.2	View details →
Adaptive Task Routing哲儀朱Recommend what to do next at task boundaries: keep, hand off, or change AI settings when worthwhile. Publisher keywords · listing agent-skills context-routing model-routing reasoning-effort agentic-workflows Show in context → 1 more matching sources Publisher full description …tings and upgrade rationale. Ordinary changes ask only about the target setting. Actual savings and reliability improvements require evidence. Show in context →	ProductivityVersion 0.5.0	View details →
Poka-YokerainmanjamMistake-proof your code, your pipeline, and your agents. Audits code for mistakes that are possible, designs APIs where misuse cannot be ... Publisher keywords · listing …of invalid states unrepresentable lean mistake proof mistake-proofing poka-yoke poke yoke poke-yoke reliability shigeo shingo shingo toyota production system type-safety zero defects Show in context → 1 more matching sources Publisher description Mistake-proof your code, your pipeline, and your agents. Audits code for mistakes that are possible, designs APIs where misuse cannot be expressed, installs guardrails, and turns incidents into devices that prevent recurrence. Show in context →	Developer ToolsVersion 0.2.0	View details →
GraffiticodeArtcompiler Inc Publisher description Graffiticode lets AI agents perform verified, task-specific operations instead of making unrestricted API calls. By using formal task languages to validate inputs, enforce capabilities, and produce structured artifacts, Graffiticode helps developers expose their services to AI agents with greater control, reliability, and predictable results. Explore the growing catalog of Graffiticode languages for common tasks, or create your own language to expose your application's unique capabilities to AI agents… Show in context →	Developer ToolsVersion 1.0.0	View details →
RoseyInSummary Publisher description …sibility — to surface weaknesses before an output reaches its intended audience. Why it matters for agentic workflows: an agent that generates important outputs for users or agents downstream cannot reliably verify its own output. Verification benefits from a separate system checking claims against… …eating important output that will be consumed as inputs by another agent in a workflow and requires reliability. Not for: scenarios where there is not an underlying source-of-truth that can be provided directly to Rosey. Show in context →	Business & OperationsVersion 1.0.0	View details →