claude-code

Plugin claude-code from marketplace rdl-agent-extensions in nq-rdl/agent-extensions. Commands read from the repo's manifest — 27 days ago.

Skills (9)

• agent-create

  Create or import a delegatable workflow into this agent-extensions catalog. Use when adding or adapting an agent procedure: author a task-focused skill, place the optional subagent outline in references/subagent.rst, and complete registry, generated plugin, changelog, and validation wiring. repo: ht

• agent-teams

  Design and run Claude Code agent teams: parallel teammates with a shared task list who message and challenge each other. Use when the user asks for a team or teammates, wants parallel reviewers or investigators who debate findings, asks whether to use a team or subagents, or mentions teammate mode,

• create-workflow

  Create, save, and reuse Claude Code dynamic workflows — JavaScript orchestration scripts that Claude writes and a runtime executes to fan out to many subagents at scale. Use when the user wants to build a workflow, automate a repeatable multi-agent process, codify an orchestration, or save a run for

• discover-plugins

  Recommend Claude Code plugins for a repository from the team's tracked marketplaces. Use when choosing plugins during Claude Code setup or when asked which plugins a project should use. Matches live marketplace catalogs to the repository's stack. Recommends only; never installs plugins or edits sett

• hook

  Create, debug, and audit Claude Code hooks in settings.json, plugins, or skill frontmatter. Use for PreToolUse, PostToolUse, PermissionRequest, Stop, SessionStart, or UserPromptSubmit hooks; command blocking, auto-formatting, auto-approval, context injection, or completion checks; or when hook outpu

• pipeline

  Design reproducible multi-step flows in Claude Code — decide between a single ordered skill, multiple step commands, hook-gated state, and when to escalate a step to subagents, an agent team, or a dynamic workflow. Use when the user wants a repeatable procedure (step 1 → 2 → 3), a "pipeline", enforc

• skill-audit

  Audit a skill for non-inferable value. Use when creating, compressing, or reviewing a SKILL.md to check it encodes a gap the fresh model cannot see rather than restating public specs. Applies the Biggs "could a fresh model write this verbatim?" test and the SkillsBench conciseness rubric, checks ver

• skill-report-issue

  Report a defect in an installed skill to the skill's upstream GitHub repository. Use when a skill gives wrong instructions, errors, or fails silently, or when the user says "this skill is broken", "file a bug for this skill", or "report this to the skill author". Searches for duplicates, drafts one

• skill-review

  Review skills changed during a development session for actionable bugs and improvements. Use after creating, editing, or debugging a skill to capture user corrections and close the feedback loop; optionally delegate a second opinion using the referenced worker outline. repo: https://github.com/nq-rd

Hooks (1)

• PostToolUse

  "${CLAUDE_PLUGIN_ROOT}/scripts/skill-audit-nudge.sh"

Similar plugins