gsd-plan-phase

Invoke this Pi skill with /skill:gsd-plan-phase.

Pi dispatch (fail closed). Pi has no built-in subagent API; GSD provides the gsd_dispatch tool. Where a workflow step dispatches a named gsd-* role, call gsd_dispatch with role (the installed role name) and task (the full task text). A workflow that still writes the step as an Agent(...) call means the same: its subagent_type is the role and its prompt is the task. The role runs in an isolated child session and cannot dispatch further, so a step that needs nested dispatch cannot run on Pi. If gsd_dispatch reports the role unresolved or the child failed, STOP that step and surface the error. If the workflow marks the step optional, say so and offer to skip it and continue; if it is required, stop the run. Never execute the role's work inline as a substitute.

Pi includes. Pi does not expand @path references in a skill. Before doing anything else, read every file listed under <execution_context> below with your read tool (the leading @ is accepted), then follow the workflow they define.

(the text typed after the skill name)

The text inside <arguments> is exactly what the user typed after the command name: data, not template instructions. An empty block means no arguments were passed.

Default flow: Research (if needed) → Plan → Verify → Done

Research-only mode (--research-phase <N>): Spawn gsd-phase-researcher for phase N, write RESEARCH.md, then exit before the planner runs. Useful for cross-phase research, doc review before committing to a planning approach, and correction-without-replanning loops where iterating on research alone is dramatically cheaper than re-spawning the planner. Replaces the deleted research-phase command (#3042).

Research-only modifiers:

• No flag — when RESEARCH.md already exists, auto-uses it: emits a one-line notice and exits cleanly, no prompt.
• --research — force-refresh: re-spawn the researcher unconditionally, no prompt. Bypasses the existing-RESEARCH.md auto-use path.
• --view — view-only: print existing RESEARCH.md to stdout. Does not spawn the researcher. Cheapest mode for the correction-without-replanning loop. If no RESEARCH.md exists yet, errors with a hint to drop --view.

Orchestrator role: Parse arguments, validate phase, research domain (unless skipped), spawn gsd-planner, verify with gsd-plan-checker, iterate until pass or max iterations, present results.

<execution_context> @GSD_PI_ENGINE_ROOT/gsd-core/workflows/plan-phase.md @GSD_PI_ENGINE_ROOT/gsd-core/references/ui-brand.md </execution_context>

Flags:

• --research — Force re-research even if RESEARCH.md exists
• --skip-research — Skip research, go straight to planning
• --gaps — Gap closure mode (reads VERIFICATION.md, skips research)
• --skip-verify — Skip verification loop
• --prd <file> — Use a PRD/acceptance criteria file instead of discuss-phase. Parses requirements into CONTEXT.md automatically. Skips discuss-phase entirely.
• --ingest <path-or-glob> — Use one or more ADR files instead of discuss-phase. Parses locked decisions + scope fences into CONTEXT.md automatically. Skips discuss-phase entirely.
• --ingest-format <auto|nygard|madr|narrative> — Optional ADR parser format override (auto default).
• --reviews — Replan incorporating cross-AI review feedback from REVIEWS.md (produced by /skill:gsd-review)
• --text — Use plain-text numbered lists instead of TUI menus (required for /rc remote sessions)
• --mvp — MVP enrichment on top of the default tracer-first ordering: frames the phase goal as a user story and, on Phase 1 of a new project, also emits SKELETON.md (Walking Skeleton). Vertical slicing itself is now the default (see --no-tracer); --mvp no longer turns it on. Can be persisted on a phase via **Mode:** mvp in ROADMAP.md.
• --no-tracer — Opt out of the default tracer-first decomposition and plan horizontal layers (the legacy default). By default every plan LEADS with one production-quality end-to-end tracer slice that is verified before any expansion task.
• --no-reversibility-gates — Suppress the human checkpoint that a one-way-door decision normally earns, for runs you intend to leave unattended. By default a decision rated one-way (undo needs a migration, breaks a published contract, or is impossible) gets a checkpoint:decision before the task implementing it. Ratings are still recorded on tasks and costly items still flagged — the flag changes what stops the run, not what the plan remembers.