bishx

Plugin bishx in bish-x/claude-bishx. Commands read from the repo's manifest — 1 month ago.

Skills (6)

• init-sync

  Scan codebase and fill/update CLAUDE.md and AGENTS.md with real project data. Produces deep architectural documentation matching production quality.

• plan

  Use for complex features needing bulletproof plans. Automated 10-actor verification pipeline (Researcher → Planner → [Skeptic + TDD + Completeness + Integration + Security* + Performance*] → Critic) that iterates up to 10 times until the Critic approves (≥75%), producing a one-shot-ready plan.

• prompt

  Generates a structured planning prompt for bishx-plan. Analyzes the user's idea, discovers relevant skills, and builds a detailed prompt for execution.

• research

  Deep research pipeline for thorough investigation of any topic. Multi-agent system with Scout, Investigators, Deep Diver, Verifier, Synthesizer, and iterative Critic review. Produces verified, sourced research reports.

• run

  Execute bd tasks with Agent Teams. Lead → Dev → 3 Reviewers (Bug + Security + Compliance) → Validate → QA.

• site

  Full website audit — UX/UI design, marketing, SEO, accessibility, performance, conversion. Crawls the site (up to 100 pages) via cmux browser, generates detailed visual critique with actionable recommendations.

Slash commands (13)

• /bishx:cancel

  Stop the active bishx-run or bishx-site session gracefully

• /bishx:idea

  Analyze the project and suggest new features worth building

• /bishx:init-sync

  Scan codebase and fill/update CLAUDE.md and AGENTS.md with real project data

• /bishx:init

  Create CLAUDE.md and AGENTS.md templates in the project root

• /bishx:plan-to-bd-tasks

  Import bishx-plan into bd (beads) as Epic → Feature → Task hierarchy with dependencies. Handles existing tasks gracefully.

• /bishx:plan

  Start automated iterative planning with 10-actor parallel verification pipeline (up to 10 iterations)

• /bishx:polish

  Analyze the project and suggest optimizations, refactors, and technical improvements

• /bishx:prompt

  Generate a structured planning prompt from a raw idea. Analyzes codebase, discovers skills, builds detailed prompt for bishx-plan.

• /bishx:research

  Deep research pipeline for thorough investigation of any topic. Multi-agent system with Scout, Investigators, Deep Diver, Verifier, Synthesizer, and iterative Critic review.

• /bishx:run

  Execute bd tasks with multi-agent orchestration (Lead → Dev → Reviewer → QA)

• /bishx:site

  Full website audit — UX/UI design, marketing, SEO, accessibility, performance, conversion. Crawls every page via cmux browser, generates detailed visual critique with actionable recommendations.

• /bishx:status

  Show current bishx-run session status

• /bishx:test

  Deep system testing — auto-detects stack, discovers components, runs unit/integration/E2E/security/performance tests, reports bugs to bd with full reproduction steps.

Subagents (16)

• completeness-validator

  Requirements traceability specialist for bishx-plan. Systematically maps every requirement from CONTEXT.md to plan tasks. Finds orphaned requirements and scope creep.

• critic

  Final quality gate for bishx-plan. Aggregates all actor feedback, scores with weighted dimensions, validates against cross-actor ceilings, and issues APPROVED/REVISE/REJECT verdict.

• dry-run-simulator

  Plan executability validator for bishx-plan. Mentally executes the first tasks as a fresh Claude session and reports where it gets stuck.

• integration-validator

  Inter-task compatibility specialist for bishx-plan. Validates data flow between tasks, dependency graph integrity, shared resource conflicts, and interface contracts.

• performance-auditor

  Performance anti-pattern detection specialist for bishx-plan. Checks for N+1 queries, missing indexes, algorithm complexity, memory leaks, and caching opportunities.

• planner

  Creates bite-sized, TDD-embedded, one-shot-executable implementation plans for bishx-plan. Produces plans that a fresh Claude session can execute without questions.

• research-critic

  Persistent critical review agent for bishx-research. Performs iterative multi-level review of research drafts, finding mirages, gaps, coherence issues, depth problems, and bias.

• research-deep-diver

  Opus-level deep dive agent for bishx-research. Goes deeper on complex areas, resolves contradictions, and verifies low-confidence critical findings from Wave 1 investigators.

• research-investigator

  Deep investigation agent for bishx-research. Researches assigned sub-topics with embedded skepticism, source verification, and multi-tier confidence tagging.

• research-scout

  Rapid landscape reconnaissance for bishx-research. Maps sources, classifies complexity, recommends investigation strategy before deep research begins.

• research-synthesizer

  Synthesis and report writing specialist for bishx-research. Combines verified findings into a structured, coherent research report organized by analytical lenses.

• research-verifier

  Fact-checking specialist for bishx-research. Verifies every claim against its cited source, detects mirages, resolves confidence inflation, and validates source accessibility.

• researcher

  Deep research agent for bishx-plan. Investigates codebase architecture, external docs, APIs, schemas, and library compatibility with maximum depth and confidence tagging.

• security-reviewer

  Security vulnerability detection specialist for bishx-plan. Checks OWASP Top 10, threat modeling, auth boundaries, input validation, and sensitive data handling.

• skeptic

  Mirage detection specialist for bishx-plan. Verifies plan claims against codebase reality and external facts. Catches assumptions masquerading as facts.

• tdd-reviewer

  TDD compliance reviewer for bishx-plan. Ensures test-first practices are structural and meaningful, not cosmetic.

Hooks (1)

• Stop

  ${CLAUDE_PLUGIN_ROOT}/hooks/stop-hook.sh