pleaselisted

# please Use this skill to run the full PRD pipeline from the current conversation to a finished implementation in a single invocation, with no human approval round-trips except for risky work. This skill is an orchestrator, not a new pipeline. It chains the existing `gen-prd`, `implement`, and `ship` skills and their harnesses exactly as written. Where this document is silent, the chained skill's own rules apply unchanged. Read `~/.codex/skills/gen-prd/SKILL.md`, `~/.codex/skills/implement/SKILL.md`, and (when delivery mode is `pr`) `~/.codex/skills/ship/SKILL.md` before executing their stages. Match the user's language by default. ## What This Skill Changes Versus The Manual Chain Only three things differ from running the skills by hand: 1. The PRD source is the current conversation, not a required intake artifact. 2. The human PRD-approval round-trip is replaced by recording the user's `$please` invocation as the approval deviation. 3. The stage transitions (PRD ready -> implement -> ship) happen automatically instead of waiting for the user to invoke the next skill. Everything else stays identical: harness state files, review profiles, audits, verification evidence, receipts, delivery config, and hard stops. Do not weaken any gate because this is the automated path. ## Inputs The requirement source is the current conversation. An argument after `$please` is a topic brief or emphasis, not a replacement for the conversation. Before starting, capture verbatim the u