Files
msd-core/get-shit-done/workflows/resume-project.md
Quang Do d4767ac2e0 fix: replace /gsd: slash command format with /gsd- skill format in all user-facing content (#1579)
* fix: replace /gsd: command format with /gsd- skill format in all suggestions

All next-step suggestions shown to users were still using the old colon
format (/gsd:xxx) which cannot be copy-pasted as skills. Migrated all
occurrences across agents/, commands/, get-shit-done/, docs/, README files,
bin/install.js (hardcoded defaults for claude runtime), and
get-shit-done/bin/lib/*.cjs (generate-claude-md templates and error messages).
Updated tests to assert new hyphen format instead of old colon format.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* fix: migrate remaining /gsd: format to /gsd- in hooks, workflows, and sdk

Addresses remaining user-facing occurrences missed in the initial migration:

- hooks/: fix 4 user-facing messages (pause-work, update, fast, quick)
  and 2 comments in gsd-workflow-guard.js
- get-shit-done/workflows/: fix 21 Skill() literal calls that Claude
  executes directly (installer does not transform workflow content)
- sdk/prompt-sanitizer.ts: update regex to strip /gsd- format in addition
  to legacy /gsd: format; update JSDoc comment
- tests/: update autonomous-ui-steps, prompt-sanitizer to assert new format

Note: commands/gsd/*.md frontmatter (name: gsd:xxx) intentionally unchanged
— installer derives skillName from directory path, not the name field.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* fix(plan-phase): preserve --chain flag in auto-advance sync and handle ui-phase gate in chain mode

Bug 1: step 15 sync-flag check only guarded against --auto, causing
_auto_chain_active to be cleared when plan-phase is invoked without
--auto in ARGUMENTS even though a --chain pipeline was active. Added
--chain to the guard condition, matching discuss-phase behaviour.

Bug 2: UI Design Contract gate (step 5.6) always exited the workflow
when UI-SPEC was missing, breaking the discuss --chain pipeline
silently. When _auto_chain_active is true, the gate now auto-invokes
gsd-ui-phase --auto via Skill() and continues to step 6 without
prompting. Manual invocations retain the existing AskUserQuestion flow.

* fix: remove <sub>/clear</sub> pattern and duplicate old-format command in discuss-phase.md

---------

Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
2026-04-04 07:24:31 -04:00

9.9 KiB

Use this workflow when: - Starting a new session on an existing project - User says "continue", "what's next", "where were we", "resume" - Any planning operation when .planning/ already exists - User returns after time away from project Instantly restore full project context so "Where were we?" has an immediate, complete answer.

<required_reading> @~/.claude/get-shit-done/references/continuation-format.md </required_reading>

Load all context in one call:
INIT=$(node "$HOME/.claude/get-shit-done/bin/gsd-tools.cjs" init resume)
if [[ "$INIT" == @file:* ]]; then INIT=$(cat "${INIT#@file:}"); fi

Parse JSON for: state_exists, roadmap_exists, project_exists, planning_exists, has_interrupted_agent, interrupted_agent_id, commit_docs.

If state_exists is true: Proceed to load_state If state_exists is false but roadmap_exists or project_exists is true: Offer to reconstruct STATE.md If planning_exists is false: This is a new project - route to /gsd-new-project

Read and parse STATE.md, then PROJECT.md:

cat .planning/STATE.md
cat .planning/PROJECT.md

From STATE.md extract:

  • Project Reference: Core value and current focus
  • Current Position: Phase X of Y, Plan A of B, Status
  • Progress: Visual progress bar
  • Recent Decisions: Key decisions affecting current work
  • Pending Todos: Ideas captured during sessions
  • Blockers/Concerns: Issues carried forward
  • Session Continuity: Where we left off, any resume files

From PROJECT.md extract:

  • What This Is: Current accurate description
  • Requirements: Validated, Active, Out of Scope
  • Key Decisions: Full decision log with outcomes
  • Constraints: Hard limits on implementation
Look for incomplete work that needs attention:
# Check for structured handoff (preferred — machine-readable)
cat .planning/HANDOFF.json 2>/dev/null || true

# Check for continue-here files (mid-plan resumption)
ls .planning/phases/*/.continue-here*.md 2>/dev/null || true

# Check for plans without summaries (incomplete execution)
for plan in .planning/phases/*/*-PLAN.md; do
  [ -e "$plan" ] || continue
  summary="${plan/PLAN/SUMMARY}"
  [ ! -f "$summary" ] && echo "Incomplete: $plan"
done 2>/dev/null || true

# Check for interrupted agents (use has_interrupted_agent and interrupted_agent_id from init)
if [ "$has_interrupted_agent" = "true" ]; then
  echo "Interrupted agent: $interrupted_agent_id"
fi

If HANDOFF.json exists:

  • This is the primary resumption source — structured data from /gsd-pause-work
  • Parse status, phase, plan, task, total_tasks, next_action
  • Check blockers and human_actions_pending — surface these immediately
  • Check completed_tasks for in_progress items — these need attention first
  • Validate uncommitted_files against git status — flag divergence
  • Use context_notes to restore mental model
  • Flag: "Found structured handoff — resuming from task {task}/{total_tasks}"
  • After successful resumption, delete HANDOFF.json (it's a one-shot artifact)

If .continue-here file exists (fallback):

  • This is a mid-plan resumption point
  • Read the file for specific resumption context
  • Flag: "Found mid-plan checkpoint"

If PLAN without SUMMARY exists:

  • Execution was started but not completed
  • Flag: "Found incomplete plan execution"

If interrupted agent found:

  • Subagent was spawned but session ended before completion
  • Read agent-history.json for task details
  • Flag: "Found interrupted agent"
Present complete project status to user:
╔══════════════════════════════════════════════════════════════╗
║  PROJECT STATUS                                               ║
╠══════════════════════════════════════════════════════════════╣
║  Building: [one-liner from PROJECT.md "What This Is"]         ║
║                                                               ║
║  Phase: [X] of [Y] - [Phase name]                            ║
║  Plan:  [A] of [B] - [Status]                                ║
║  Progress: [██████░░░░] XX%                                  ║
║                                                               ║
║  Last activity: [date] - [what happened]                     ║
╚══════════════════════════════════════════════════════════════╝

[If incomplete work found:]
⚠️  Incomplete work detected:
    - [.continue-here file or incomplete plan]

[If interrupted agent found:]
⚠️  Interrupted agent detected:
    Agent ID: [id]
    Task: [task description from agent-history.json]
    Interrupted: [timestamp]

    Resume with: Task tool (resume parameter with agent ID)

[If pending todos exist:]
📋 [N] pending todos — /gsd-check-todos to review

[If blockers exist:]
⚠️  Carried concerns:
    - [blocker 1]
    - [blocker 2]

[If alignment is not ✓:]
⚠️  Brief alignment: [status] - [assessment]
Based on project state, determine the most logical next action:

If interrupted agent exists: → Primary: Resume interrupted agent (Task tool with resume parameter) → Option: Start fresh (abandon agent work)

If HANDOFF.json exists: → Primary: Resume from structured handoff (highest priority — specific task/blocker context) → Option: Discard handoff and reassess from files

If .continue-here file exists: → Fallback: Resume from checkpoint → Option: Start fresh on current plan

If incomplete plan (PLAN without SUMMARY): → Primary: Complete the incomplete plan → Option: Abandon and move on

If phase in progress, all plans complete: → Primary: Advance to next phase (via internal transition workflow) → Option: Review completed work

If phase ready to plan: → Check if CONTEXT.md exists for this phase:

  • If CONTEXT.md missing: → Primary: Discuss phase vision (how user imagines it working) → Secondary: Plan directly (skip context gathering)
  • If CONTEXT.md exists: → Primary: Plan the phase → Option: Review roadmap

If phase ready to execute: → Primary: Execute next plan → Option: Review the plan first

Present contextual options based on project state:
What would you like to do?

[Primary action based on state - e.g.:]
1. Resume interrupted agent [if interrupted agent found]
   OR
1. Execute phase (/gsd-execute-phase {phase} ${GSD_WS})
   OR
1. Discuss Phase 3 context (/gsd-discuss-phase 3 ${GSD_WS}) [if CONTEXT.md missing]
   OR
1. Plan Phase 3 (/gsd-plan-phase 3 ${GSD_WS}) [if CONTEXT.md exists or discuss option declined]

[Secondary options:]
2. Review current phase status
3. Check pending todos ([N] pending)
4. Review brief alignment
5. Something else

Note: When offering phase planning, check for CONTEXT.md existence first:

ls .planning/phases/XX-name/*-CONTEXT.md 2>/dev/null || true

If missing, suggest discuss-phase before plan. If exists, offer plan directly.

Wait for user selection.

Based on user selection, route to appropriate workflow:
  • Execute plan → Show command for user to run after clearing:
    ---
    
    ## ▶ Next Up
    
    **{phase}-{plan}: [Plan Name]** — [objective from PLAN.md]
    
    `/clear` then:
    
    `/gsd-execute-phase {phase} ${GSD_WS}`
    
    ---
    
  • Plan phase → Show command for user to run after clearing:
    ---
    
    ## ▶ Next Up
    
    **Phase [N]: [Name]** — [Goal from ROADMAP.md]
    
    `/clear` then:
    
    `/gsd-plan-phase [phase-number] ${GSD_WS}`
    
    ---
    
    **Also available:**
    - `/gsd-discuss-phase [N] ${GSD_WS}` — gather context first
    - `/gsd-research-phase [N] ${GSD_WS}` — investigate unknowns
    
    ---
    
  • Advance to next phase → ./transition.md (internal workflow, invoked inline — NOT a user command)
  • Check todos → Read .planning/todos/pending/, present summary
  • Review alignment → Read PROJECT.md, compare to current state
  • Something else → Ask what they need
Before proceeding to routed workflow, update session continuity:

Update STATE.md:

## Session Continuity

Last session: [now]
Stopped at: Session resumed, proceeding to [action]
Resume file: [updated if applicable]

This ensures if session ends unexpectedly, next resume knows the state.

If STATE.md is missing but other artifacts exist:

"STATE.md missing. Reconstructing from artifacts..."

  1. Read PROJECT.md → Extract "What This Is" and Core Value
  2. Read ROADMAP.md → Determine phases, find current position
  3. Scan *-SUMMARY.md files → Extract decisions, concerns
  4. Count pending todos in .planning/todos/pending/
  5. Check for .continue-here files → Session continuity

Reconstruct and write STATE.md, then proceed normally.

This handles cases where:

  • Project predates STATE.md introduction
  • File was accidentally deleted
  • Cloning repo without full .planning/ state

<quick_resume> If user says "continue" or "go":

  • Load state silently
  • Determine primary action
  • Execute immediately without presenting options

"Continuing from [state]... [action]" </quick_resume>

<success_criteria> Resume is complete when:

  • STATE.md loaded (or reconstructed)
  • Incomplete work detected and flagged
  • Clear status presented to user
  • Contextual next actions offered
  • User knows exactly where project stands
  • Session continuity updated </success_criteria>