Skill Codex
Create product-facing documents only when explicitly requested: PRDs, user scenarios, baselines, feature impact, roadmaps, stakeholder packages, and product implementation or hardening plans. Do not use for code changes, approved-plan execution, technical research, benchmarks, or developer-tool implementation unless…
Skill Codex
Use when bootstrapping, refreshing, auditing, pruning, or promoting durable knowledge for one service repository; when onboarding docs are missing or fragmented; or when topology, entrypoints, verification commands, integrations, risk zones, or operational knowledge changed. Not for ordinary feature work unless a…
Skill Claude CodeCodex
A meeting-transcript tool that extracts explicit tasks and commitments into structured JSON, including the owner, deadline, and exact supporting quote.
Skill Claude CodeCodex
A meeting-transcript tool that extracts explicit tasks and commitments into structured JSON, including the owner, deadline, and exact supporting quote.
Skill Claude CodeCodex
A meeting-transcript tool that extracts explicit tasks and commitments into structured JSON, including the owner, deadline, and exact supporting quote.
Skill Claude CodeCodex
An unfinished skill for extracting a deadline from a Russian work note and preserving the exact quoted wording. It is marked blocked and is not ready for use.
Skill Codex
Create and improve agent skills by resolving consequential requirements, defining acceptance examples, running evaluations, and adding an evidence-based debug mode. Use when a user wants a working, testable skill or needs to diagnose a skill's failures; not for ordinary execution of an unrelated skill.
Skill Codex
Create and improve agent skills by resolving consequential requirements, defining acceptance examples, running evaluations, and adding an evidence-based debug mode. Use when a user wants a working, testable skill or needs to diagnose a skill's failures; not for ordinary execution of an unrelated skill.
Skill Codex
Create and improve agent skills by resolving consequential requirements, defining acceptance examples, running evaluations, and adding an evidence-based debug mode. Use when a user wants a working, testable skill or needs to diagnose a skill's failures; not for ordinary execution of an unrelated skill.
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: