tea-rags

Plugin tea-rags from marketplace tea-rags in artk0de/TeaRAGs-MCP. Commands read from the repo's manifest — 2 days ago.

Skills (17)

• analytics-rerank

  Pick tea-rags rerank preset, build `{custom: {...}}` weight set, or compose analytics-driven search (ownership, tech-debt, hotspots, code-review, security-audit, fragile-silo). Invoke on intent like "rank by ownership", "find tech debt", "hotspot scan", "code review of recent changes", "security aud

• architecture-diagnostics

  Diagnose structural violations for an explicit architecture audit, an explicit invocation of this skill, or structural refactoring. Assess dependency/layer violations, coupling and move candidates. Not for context gathering, code overviews, behavior or contract questions, or routine feature work. Ri

• bug-hunt

  Find source of concrete failure — rank historically buggy code (high bugFixRate + churn) AND freshly changed never-fixed code (committed or still uncommitted) against symptom. Triggers on "debug X", "why does Y fail", "test fails", "stack trace says Z", "find bugs in X", "found a bug in X — show / c

• data-driven-generation

  Decide HOW to write or change code by inspecting git signals on neighbors — classifies task mode (CREATE / EXTEND / MODIFY), pulls strategy and templates from proven low-churn implementations, reuses existing shared helpers. Triggers on "implement function X", "add method to class Y", "write a helpe

• explore

  Use for questions about existing code behavior, usages, structure, examples, or explicit pre-change context that need investigation. Not bug symptoms, implementation-plan authoring, or broad code-health audits.

• extract-project-patterns

  Agentic-only — runs on any behaviorQuery / positiveIds handoff: proven code as code-gen templates, not test patterns. Returns templates + locality (target subdomain → infra prefix → project). Read body before planning. Called by data-driven-generation Step 2, dinopowers writing-plans / executing-pla

• filter-building

  Build tea-rags filter beyond `pathPattern`. Invoke whenever reasoning before a tea-rags search includes a SCOPE: domain, language, author, time window, ticket, prod-vs-test split, directory exclusion. User names scope ("in the X domain", "Ruby code", "Alice's work", "modified this week", "for ticket

• force-reindex

  Force full re-index, zero downtime: new versioned collection built in background while search continues; alias switches atomically. Also for a corrupt index. Requires explicit user confirmation — NEVER invoke automatically.

• index

  Index/reindex codebase: new → alias + index (index-codebase --name); indexed → incremental; corrupt → force-reindex. NOT status checks (get_index_status).

• mr-review

  Signal-driven review of merge request or local branch — rank diff risks via tea-rags (blast radius, fragile zones, silo style, missing tests, doc invariants, cycles, naming), emit plain-language comments each naming a concrete fix. Triggers on "review this MR <url>", "review MR/PR", "проведи ревью M

• pattern-search
• refactoring-planning

  Turn architecture diagnosis into an ordered, simulated refactoring plan — restructure ops per finding, ordered by tier, handed to writing-plans as Tasks. Use when asked "plan a refactoring", "plan the architecture fix", "fix order for these violations", "refactoring roadmap", "спланируй рефакторинг"

• refactoring-scan
• report-issue

  Help the user file a high-quality GitHub issue for a TeaRAGs bug — gather environment + diagnostic context, CHECK FOR EXISTING/KNOWN ISSUES FIRST, then produce a pre-filled issue URL (or `gh issue create` when available). Triggers on "report a bug", "file an issue", "open a GitHub issue about this e

• risk-assessment

  Assess project/domain health — multi-preset risk scan (bugs, hotspots, tech debt) into tiers. Use to evaluate risks, assess code health, find zones needing attention — NOT one-preset asks like bug-prone areas (analytics-rerank), NOT bug symptoms (bug-hunt) or an unframed "something's off in X" (dino

• test-first

  Agentic-only RED-phase skill — writes failing tests BEFORE implementation, from spec/plan acceptance criteria or given test code verbatim, else one test per behaviour of the task. Runs them; each must fail for the RIGHT reason (assertion / missing behaviour, not import / compile / fixture errors). M

• tests-as-context

  Agentic-only — DSL test chunks ("test" scenarios, "test_setup" fixtures) as context for review, verification, refactoring, debugging; a user asking for test patterns / conventions wants the dinopowers TDD wrapper, not this recipe. NOT user-invocable: a user naming this skill or asking which tests co

Slash commands (1)

• /tea-rags:prime

Hooks (13)

• SessionStart

  ${CLAUDE_PLUGIN_ROOT}/scripts/inject-rules.sh --part 1 --parts 1

• SessionStart

  tea-rags prime --agent "$CLAUDE_PROJECT_DIR"

• SessionStart

  ${CLAUDE_PLUGIN_ROOT}/scripts/check-build-fresh.sh

• SessionStart

  ${CLAUDE_PLUGIN_ROOT}/scripts/reread-ledger.mjs

• UserPromptSubmit

  ${CLAUDE_PLUGIN_ROOT}/scripts/turn-routing.sh --harness claude

• UserPromptSubmit

  ${CLAUDE_PLUGIN_ROOT}/scripts/search-drift.sh --harness claude

• UserPromptSubmit

  ${CLAUDE_PLUGIN_ROOT}/scripts/reread-ledger.mjs

• PreToolUse

  ${CLAUDE_PLUGIN_ROOT}/scripts/search-drift.sh --harness claude

• PreToolUse

  ${CLAUDE_PLUGIN_ROOT}/scripts/reread-ledger.mjs

• PostToolUse

  ${CLAUDE_PLUGIN_ROOT}/scripts/search-drift.sh --harness claude

• PostToolUse

  ${CLAUDE_PLUGIN_ROOT}/scripts/cleanup-worktree-clone.sh

• PostToolUse

  ${CLAUDE_PLUGIN_ROOT}/scripts/reread-ledger.mjs

• SubagentStart

  ${CLAUDE_PLUGIN_ROOT}/scripts/enforce-tearags-search.sh

Similar plugins