synthex

Plugin synthex from marketplace lumenai in bluminal/lumenai. Commands read from the repo's manifest — yesterday.

Skills (47)

• architect

  Reviews system architecture, feasibility, and technical trade-offs; writes ADRs and RFC sections.

• bedrock-review-prompter

  Adapter that invokes AWS Bedrock as an external code-review proposer in multi-model review.

• cancel-loop

  Mark a running native-looping loop cancelled; polled at the next iteration boundary. Idempotent.

• claude-review-prompter

  Specialty adapter invoking a second Claude CLI session as an external multi-model review proposer.

• code-reviewer

  Reviews code for craftsmanship, correctness, convention adherence, and reuse opportunities.

• codex-review-prompter

  Adapter that invokes the OpenAI Codex CLI as an external proposer in multi-model review.

• configure-multi-model

  Wizard to enable or reconfigure multi-model code review (fan out to OpenAI, Google, Ollama, etc.).

• configure-teams

  Wizard to enable or reconfigure standing review pool routing (routing_mode, matching_mode).

• design-system-agent

  Owns the design token registry and component library; audits frontend design-system compliance.

• design-system-audit

  Audit the frontend for design-token violations, misused components, and accessibility issues.

• dismiss-upgrade-nudge

  Silence the Synthex SessionStart upgrade nudge for this project until manually re-enabled.

• findings-consolidator

  Deduplicates and merges findings from multiple reviewers into one attributed, sorted list.

• gemini-review-prompter

  Adapter that invokes the Gemini CLI as an external code-review proposer in multi-model review.

• init

  Scaffold .synthex/config.yaml and the docs/reqs, docs/plans, docs/specs directories for a new project.

• lead-frontend-engineer

  Leads frontend delivery: UI implementation, UX quality, accessibility, and design-system use.

• list-loops

  List running and recent native-looping loops tracked in .synthex/loops/. Read-only.

• list-teams

  List active standing pools and non-standing per-invocation teams with roster, tasks, TTL.

• llm-review-prompter

  Adapter that invokes Simon Willison's llm CLI as an external code-review proposer.

• loop

  Run an arbitrary prompt iteratively in this thread until a completion promise or max-iterations.

• metrics-analyst

  Reports DORA, HEART, and AARRR metrics and tracks OKR progress from engineering data.

• multi-model-review-orchestrator

  Fans a review out to native and external CLI proposers in parallel and consolidates findings.

• next-priority

  Execute the next highest-priority implementation-plan tasks via the Tech Lead sub-agent.

• ollama-review-prompter

  Adapter that POSTs to a local Ollama server as an external code-review proposer.

• performance-audit

  Run a full-stack performance audit quantifying bottlenecks against budgets with fixes.

• performance-engineer

  Analyzes full-stack performance: Core Web Vitals, queries, and bundles, with quantified fixes.

• product-manager

  Gathers requirements and turns them into a prioritized, value-driven implementation plan.

• quality-engineer

  Designs test strategy, writes test suites, and analyzes coverage gaps across the codebase.

• refine-requirements

  Run a PRD through multi-agent review for clarity and completeness, then apply the edits.

• reliability-review

  Assess a service's operational readiness: SLOs, observability, runbooks, deployment risk.

• retrospective-facilitator

  Runs a structured retrospective and produces a bounded list of actionable improvement items.

• retrospective

  Facilitate a structured end-of-cycle retrospective combining metrics with qualitative findings.

• review-code

  Run craftsmanship, security, and optional performance review on a diff or file set in parallel.

• security-reviewer
• sre-agent
• standing-pool-cleanup
• standing-pool-submitter
• star
• start-review-team
• stop-review-team
• tech-lead
• technical-writer
• terraform-plan-reviewer
• test-coverage-analysis
• ux-researcher
• write-adr
• write-implementation-plan
• write-rfc

Slash commands (22)

• /synthex:cancel-loop
• /synthex:configure-multi-model
• /synthex:configure-teams
• /synthex:design-system-audit
• /synthex:dismiss-upgrade-nudge
• /synthex:init
• /synthex:list-loops
• /synthex:list-teams
• /synthex:loop
• /synthex:next-priority
• /synthex:performance-audit
• /synthex:refine-requirements
• /synthex:reliability-review
• /synthex:retrospective
• /synthex:review-code
• /synthex:star
• /synthex:start-review-team
• /synthex:stop-review-team
• /synthex:test-coverage-analysis
• /synthex:write-adr
• /synthex:write-implementation-plan
• /synthex:write-rfc

Subagents (25)

• architect

  Reviews system architecture, feasibility, and technical trade-offs; writes ADRs and RFC sections.

• bedrock-review-prompter

  Adapter that invokes AWS Bedrock as an external code-review proposer in multi-model review.

• claude-review-prompter

  Specialty adapter invoking a second Claude CLI session as an external multi-model review proposer.

• code-reviewer

  Reviews code for craftsmanship, correctness, convention adherence, and reuse opportunities.

• codex-review-prompter

  Adapter that invokes the OpenAI Codex CLI as an external proposer in multi-model review.

• design-system-agent

  Owns the design token registry and component library; audits frontend design-system compliance.

• findings-consolidator

  Deduplicates and merges findings from multiple reviewers into one attributed, sorted list.

• gemini-review-prompter

  Adapter that invokes the Gemini CLI as an external code-review proposer in multi-model review.

• lead-frontend-engineer

  Leads frontend delivery: UI implementation, UX quality, accessibility, and design-system use.

• llm-review-prompter

  Adapter that invokes Simon Willison's llm CLI as an external code-review proposer.

• metrics-analyst

  Reports DORA, HEART, and AARRR metrics and tracks OKR progress from engineering data.

• multi-model-review-orchestrator

  Fans a review out to native and external CLI proposers in parallel and consolidates findings.

• ollama-review-prompter

  Adapter that POSTs to a local Ollama server as an external code-review proposer.

• performance-engineer

  Analyzes full-stack performance: Core Web Vitals, queries, and bundles, with quantified fixes.

• product-manager

  Gathers requirements and turns them into a prioritized, value-driven implementation plan.

• quality-engineer

  Designs test strategy, writes test suites, and analyzes coverage gaps across the codebase.

• retrospective-facilitator

  Runs a structured retrospective and produces a bounded list of actionable improvement items.

• security-reviewer

  Reviews code for vulnerabilities, secret leaks, and access-control defects as a security gate.

• sre-agent

  Defines SLOs, designs observability, writes runbooks, and assesses deployment risk.

• standing-pool-cleanup

  Atomically removes a stale standing pool's index entry and metadata directory under the cross-session lock.

• standing-pool-submitter

  Submits review tasks to a standing pool's task list and polls for the consolidated report envelope.

• tech-lead

  Primary coding and orchestration agent; implements features and delegates to specialists.

• technical-writer

  Writes and maintains API docs, user guides, migration guides, and changelogs.

• terraform-plan-reviewer

  Reviews terraform plan output for cost, security, and destructive-change risk.

• ux-researcher

  Designs research plans and produces personas, journey maps, and Opportunity Solution Trees.

Hooks (7)

• SessionStart

  ${CLAUDE_PLUGIN_ROOT}/scripts/upgrade-nudge.sh

• SessionStart

  ${CLAUDE_PLUGIN_ROOT}/scripts/compact-recover.sh

• Stop

  ${CLAUDE_PLUGIN_ROOT}/scripts/loop-advance-gate.sh

• PreToolUse

  ${CLAUDE_PLUGIN_ROOT}/scripts/commit-lint.sh

• PreToolUse

  ${CLAUDE_PLUGIN_ROOT}/scripts/auto-decide-gate.sh

• TaskCompleted

  ${CLAUDE_PLUGIN_ROOT}/scripts/task-completed-gate.sh

• TeammateIdle

  ${CLAUDE_PLUGIN_ROOT}/scripts/teammate-idle-gate.sh

Similar plugins