validate
1201Command Claude Code
Validate one skill end to end in a clean-room agent, then report the result to the PR.
Slash commands. One file, one job, loaded with the rest of your configuration.
Command Claude Code
Validate one skill end to end in a clean-room agent, then report the result to the PR.
Command Claude Code
Math Research OS entry: auto-diagnose user intent, route to thinking lenses, math knowledge base, or design translation layer.
Command Claude Code
Create and verify a scoped Qovery API Policy Token from your intent.
Command Claude Code
Code review — local uncommitted changes or GitHub PR (pass PR number/URL for PR mode).
Command Claude Code
Archive merged feature specs into project memory with provenance, sweep discovery, and gated cleanup.
Command Claude Code
Auto-otimização com filtro de qualidade — só o que prova valor entra, tudo que não performa sai.
Command Claude Code
File a phase artifact into MemPalace; mirror decision facts into its temporal KG.
Command Claude Code
Operate as Identity Shift {OS}: a transformation operating system that converts a declared future identity into daily evidence through reflection, challenge, planning, execution, measurement, memory and selective storytelling. Do not motivate performatively.
5throck/ai-workspace-standards
Command Claude Code
Run security advisory scan (daily mode) or pre-PR advisory check (--pr flag).
Command Claude Code
A command for creating and setting up a GitHub Project, including its name, description, and custom fields, based on the repository's README file.
Command Claude Code
A command for creating a new version of a Qiniu AppMarket application. It collects version details and deployment settings, including a Terraform module, which is reusable infrastructure configuration.
Command Claude Code
A command for maintaining a project’s .context folder, which stores coding preferences, workflow notes, decision logs, and history. It creates an audit trail showing why changes were made.
Command Claude Code
A procedural document that marks stages in a document-writing process and then passes the work to a document-drafting skill.
Command Claude CodeOpenCode
Ship the current release after validation passes.
Command Claude Code
Execute a task through a cached, validated dependency graph with localized retries and a live report.
anasfik/claude-code-prompts-reference
Command Claude Code
Command "ultraplan" from anasfik/claude-code-prompts-reference, covering command prompt: /ultraplan, prompt sources, 1) default instruction text (bundled), 2) dev override (ant-only) and prompt assembly.
Command Claude Code
Run sequential code review passes with fresh contexts to catch more issues.
Command Claude Code
Execute implementation by processing atomic task files one at a time with Context Pinning (Atomic Traceability Model).
Command Claude Code
A command for expanding the latest saved project decision into a detailed explanation. It can cover the idea, how it works, alternative implementations, and reference choices.
Command Claude Code
These commands are handled by the CLI itself instead of being sent to the model.
Command Claude Code
Export this session's conversation to a markdown file.
Command Claude Code
Triple-provider adversarial review — Codex + Gemini + Claude synthesis.
Command Claude Code
End-of-session handoff — summarize, verify, and stage so you can review and commit.
nkhdiscovery/headless-agentic-codebase
Command Claude Code
Create a new numbered Architecture Decision Record.
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: