team-planex_skill

This skill orchestrates a two-person team plan-execute workflow, routing tasks by role and enabling concurrent planning and execution.
  • TypeScript

1.3k

GitHub Stars

1

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill catlog22/claude-code-workflow --skill team-planex

  • SKILL.md11.9 KB

Overview

This skill implements a two-member plan-and-execute pipeline (planner + executor) using a Wave Pipeline so planning and execution run concurrently. The planner produces PLAN-* waves and immediately spawns EXEC-* tasks; the executor polls and consumes EXEC-* tasks independently. Roles are invoked with --role and a lightweight orchestration mode spawns both agents and initial tasks.

How this skill works

The skill routes behavior by the --role argument: planner or executor. Planner waves create issue breakdowns, solution queues, and EXEC-* tasks while immediately moving to the next wave. Executor continuously polls for EXEC-* tasks, selects an execution backend (agent / codex / gemini / auto), implements solutions, runs tests, and reports back via a message bus. A team-wide message bus and strict output-tagging enforce role isolation and coordination.

When to use it

  • Coordinating parallel planning and implementation for medium-sized feature sets
  • Automating iterative delivery where planning should not block execution
  • Running CI-like implementation waves while planning continues ahead
  • Orchestrating multi-step tasks that benefit from staged handoffs
  • Testing execution backend strategies (agent vs Codex vs Gemini)

Best practices

  • Always invoke with --role when running a single role; omit --role for orchestration mode
  • Include execution_method and code_review fields in EXEC-* task descriptions to avoid ambiguity
  • Tag all outgoing messages with role-prefixed summaries (e.g., [planner], [executor]) and log via the message bus
  • Use --exec or -y to preselect execution backend for non-interactive runs
  • Keep planner responsibilities limited to planning and queueing; do not modify code from planner role

Example use cases

  • Launch a new feature: planner breaks requirements into PLAN waves while executor implements EXEC tasks concurrently
  • Large backlog sprint: planner streams waves to maintain steady executor throughput
  • Adopt different execution backends: test agent for small fixes and Codex/Gemini for complex jobs
  • Automated orchestration: call skill without --role to spawn planner/executor agents and boot the team pipeline

FAQ

The skill enters orchestration mode: it creates the team, spawns planner and executor agents, creates an initial PLAN-* task, and selects execution settings.

How is execution backend chosen?

You can explicitly set --exec=agent|codex|gemini, use -y/--auto for defaults, or allow an interactive choice; in Auto mode task count heuristics pick the backend.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational