claudecodex

cure-tri-lane

Opt-in multi-vendor orchestration: Claude architect, Codex implementation and correctness lanes (codex exec), Antigravity system-review lane (agy) in read-only worktrees, and a fresh-context advisor. Uses existing ChatGPT and Google logins; independent of cure-product-engineering.

3 on GitHubby Cure Consulting Groupv1.16.1MITupdated todayCoding
/

Plugin cure-tri-lane from marketplace cure in Cure-Consulting-Group/ProductEngineeringSkills. Commands read from the repo's manifest — today.

This plugin runs code on your machine — its hooks fire automatically on events like tool use. The exact commands are listed below; read them before installing.

What it installs

Skills (1)

  • tri-lane

    Route work across Claude (architect), Codex (implementation and correctness review), and Antigravity (system review) with a fresh-context advisor before shipping. Use when a task is big enough to delegate, touches security, migrations, API shapes, auth, billing, or CI, or has failed twice. NOT for o

Subagents (4)

  • antigravity-analyst

    Antigravity analysis lane. Runs Gemini 3.8 Flash (high) via `agy -p` in a read-only git worktree with a JSON verdict schema, on the Google AI pool. Use for system, security-rules, infra, CI, and data-model review, whole-repository reads (switch to Gemini 3.1 Pro only when 1M context is the point), a

  • codex-implementer

    Codex implementation lane. Runs GPT-6 Luna (routine) or GPT-6.1 Sol (hard) via `codex exec` in a fresh git worktree at the effort named in the spec, then returns the lane report with real verification output. Use when the architect has a finished six-part spec and wants the code written by a differe

  • codex-reviewer

    Codex correctness review lane. Runs `codex exec review` read-only against a base branch, a commit, or the uncommitted state and returns the findings verbatim, unlabelled, on the Codex Pro pool. Use on the audit route (rules, migrations, API shapes, auth, billing, CI) or after two failed attempts, fo

  • cure-advisor

    Fresh-context second opinion and final reviewer. Reads a proposed decision or a finished diff against the stated goal, with none of the conversation's accumulated assumptions, and writes the verdict to the run directory and returns it, under 300 words: ship, fix-first, or rethink. Use at commitment

Hooks (2)

  • PreToolUse

    python3 "${CLAUDE_PLUGIN_ROOT}/hooks/guard-lane-command.py"

  • SessionStart

    python3 "${CLAUDE_PLUGIN_ROOT}/skills/tri-lane/scripts/lane-log.py" due --within 1 --quiet 2>/dev/null || true

View source on GitHub

More from Cure-Consulting-Group/ProductEngineeringSkills

All 11 plugins →

Complete product engineering skill library — 104 skills (domain-organized), 41 agents (engineering, product, marketing, business, data, security), 4 personas (cross-domain engagement archetypes), 11 rules, 9 output styles, multi-layer hooks with agent auto-triggering, bundled stdlib Python tooling, LSP server configs (MCP is an opt-in template, not shipped). Standards for Android, iOS, Web, Firebase, Python, Go, Rust, Docker, Terraform, and cloud infrastructure.

104 skills · 41 agents · hooks · LSP

3

Mod: caps how many subagents one session runs at once, tracked by agent id from spawn to completion; optional research routing to a model alias, off by default. Adds /agents-budget.

3

Mod: prompt cache time left, hit rate and misses above the prompt; says what a cold next message re-caches. Adds /cache.

3

Mod: checks file:line citations in Claude's answers against the session's repositories and toasts the stale ones; never edits the answer. Adds /cite-check.

3

Mod: draws the images you sent above the prompt until the turn that carried them ends.

3

Mod: multi-lane dashboard for tri-lane worktrees, branches, live commits, and defect window status. Adds /lanes.

3

Similar plugins

Matt Pocock's agent skills for real engineering (grilling, spec/ticket flows, TDD, code review, domain modelling and more).

39 skills

mattpocock/skills 323,108

superpowers

claudecodex

Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques

15 skills · hooks

obra/superpowers 316,836

claude-mem

claudecodex

Persistent memory for Claude Code: the next session starts where the last one left off.

29 skills · 1 MCP server · hooks

thedotmack/claude-mem 116,618

career-ops

claudecodex

Open-source AI job search that runs locally in your AI coding CLI: scan job portals, evaluate listings into a structured A-H report with a 1-5 score, tailor your CV, track applications

5 skills · hooks

career-ops-hq/career-ops 82,253

mem0

claudecodex

Cross-session memory and token savings for coding agents.

6 skills · 1 agent · 1 MCP server · hooks

mem0ai/mem0 68,893

context7

claudecodex

Up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your coding agent's context.

1 MCP server

upstash/context7 64,709

Maintain cure-tri-lane?

This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.

claude & codex plugins11 plugins