paw-status

name	paw-status
description	Workflow status activity skill for PAW workflow. Diagnoses workflow state, recommends next steps, explains PAW process, and optionally posts updates to Issues/PRs.

Workflow Status

Execution Context: This skill runs directly in the PAW session (not a subagent)—simple diagnostic with no context isolation benefit.

Serve as the workflow navigator and historian. Diagnose current workflow state, recommend next actions, and optionally post updates to Issues/PRs.

Reference: Follow Core Implementation Principles from paw-workflow skill.

Capabilities

• Diagnose current workflow state from artifacts and git
• Recommend appropriate next steps
• Explain what PAW is, how the workflow works, and how to get started
• List active work items across workspace
• List available workflow presets (built-in and user-defined)
• Post status updates to Issues/PRs (on explicit request)

Default Behavior

Report status in chat—do NOT post to GitHub unless explicitly asked ("post status to issue", "update the PR summary").

State Detection

Artifact Discovery

Check .paw/work/<work-id>/ for:

• WorkflowContext.md (required for workflow context)
• Spec.md, SpecResearch.md (specification stage)
• CodeResearch.md (research stage)
• ImplementationPlan.md (planning stage)
• Docs.md (documentation if separate from implementation)
• reviews/ directory (Final Agent Review artifacts)
• reviews/planning/ directory (Planning Documents Review artifacts)

Note existence vs intentionally skipped (minimal mode skips Spec/Docs).

Configuration Detection

Read WorkflowContext.md for:

• Workflow Mode, Review Strategy, Review Policy
• Artifact Lifecycle: commit-and-clean | commit-and-persist | never-commit
• Final Agent Review: enabled | disabled
• Planning Docs Review: enabled | disabled
• Final Review Mode: single-model | multi-model | society-of-thought
• Final Review Interactive: true | false | smart
• Plan Generation Mode: single-model | multi-model
• Plan Generation Models: comma-separated model names (for multi-model modes)
• Implementation Model: none (session default) | concrete model name
• Planning Review Mode: single-model | multi-model
• Planning Review Interactive: true | false | smart
• Final Review Specialists: all | adaptive:<N> | comma-separated list (when society-of-thought)
• Final Review Interaction Mode: parallel | debate (when society-of-thought)
• Final Review Specialist Models: none | model pool | pinned pairs | mixed (when society-of-thought)

Phase Counting

Parse ImplementationPlan.md with regex: ^## Phase \d+:

• Count distinct phase numbers
• Report: "Phase N of M" or "Phase N (plan shows M phases total)"
• Never assume total—always verify

Phase Candidates

Check ## Phase Candidates section in ImplementationPlan.md:

• Count unresolved (- [ ]), promoted ([promoted]), skipped ([skipped]), deferred ([deferred]), not feasible ([not feasible])
• If unresolved exist after last phase: report "N phase candidates pending review"

Git Status

Determine:

• Current branch: git branch --show-current
• Staged/unstaged changes: git status --porcelain
• Upstream divergence: commits ahead/behind remote

PR Discovery

For prs strategy:

• Find branches: <target>_plan, <target>_phase*, <target>_docs
• Query PRs by head branch
• Capture: URL, state (open/merged/closed), CI status

For local strategy:

• Focus on target branch and Final PR only

PR Review Comments

When PRs exist:

• Fetch inline review comments and general discussion
• Compare comment timestamps vs commit history
• Summarize: "X comments (Y addressed, Z outstanding)"

Workflow Stage Progression

Map state to guidance:

State	Recommendation
Missing Spec.md + full mode	"Start specification: spec"
Spec.md exists, no CodeResearch.md	"Run code research: research"
CodeResearch.md exists, no Plan	"Create plan: plan"
Plan exists, planning-docs-review enabled, no reviews/planning/	"Run planning docs review"
Plan exists, no phase work	"Begin Phase 1: implement"
Phase N complete, Phase N+1 exists	"Continue Phase N+1: implement"
All phases complete, review enabled, no reviews/	"Run final review: final-review"
All phases complete, reviews/ exists	"Create final PR: pr"
All phases complete, review disabled	"Create final PR: pr"
All complete + unresolved candidates	"Review phase candidates before PR"

Workflow Mode Behavior

Full Mode

Expect: Spec → Spec Research (optional) → Code Research → Plan → Planning Docs Review (if enabled) → Implementation (multi-phase, including Documentation phase) → Final Review (if enabled) → Final PR

Minimal Mode

Expect: Code Research → Plan → Implementation (including Documentation phase) → Final Review (if enabled) → Final PR Skips: Spec, Spec Research (local strategy enforced)

Custom Mode

Inspect disk to discover actual stages per Custom Workflow Instructions.

Multi-Work Management

When asked "What PAW work items do I have?":

• Enumerate .paw/work/ directories with WorkflowContext.md
• Report: Work Title, Work ID, target branch, last modified, current stage
• Sort by most recently modified

Issue/PR Updates (Opt-in Only)

Only post externally when explicitly asked.

Issue comments:

• Prefix: **🐾 Status Update 🤖:**
• Include: Artifacts, PRs, phase checklist

PR body updates:

• Only modify content inside <!-- BEGIN:AGENT-SUMMARY --> / <!-- END:AGENT-SUMMARY --> block
• Preserve all other text

Never:

• Modify issue descriptions
• Assign reviewers
• Change labels

Preset Discovery

When asked "What presets are available?" or "List presets":

• Built-in presets: quick (minimal ceremony), standard (balanced with gates), thorough (maximum rigor), team (PR-based collaboration), auto (autonomous, final-pr-only), auto-full (auto + full review rigor), shaping-full (work shaping then auto-full)
• User presets: Scan ~/.paw/presets/*.yaml files, read name and description from each
• Display: name, description, source (built-in/user), default marker if set, extends target if present
• If user preset has same name as built-in, show user version (it takes precedence)

Help Mode

For "What is PAW?", "What does PAW stand for?", or similar onboarding questions:

• Use the exact name Phased Agent Workflow. Never invent an acronym expansion.
• For full documentation, see: https://lossyrob.github.io/phased-agent-workflow/ (overview), https://lossyrob.github.io/phased-agent-workflow/guide/ (getting started), https://lossyrob.github.io/phased-agent-workflow/guide/cli-installation/ (CLI install).
• Explain briefly:
  • PAW enables Context-Driven Development: agents build understanding through durable artifacts (specs, research, plans) before writing code.
  • PAW supports two workflows: implementation and review.
  • Workflow stages are top-level milestones; implementation phases are subdivisions inside ImplementationPlan.md.
• If the user also asks how to get started: {{#cli}}
  • Select the PAW custom agent (e.g. copilot --agent PAW or /agent PAW) and say "start a new PAW workflow". {{/cli}} {{#vscode}}
  • Run PAW: New PAW Workflow. {{/vscode}}
• Prefer concise, sourced onboarding language over improvisation.

When asked "What does do?", provide:

1. Purpose of the stage
2. Required inputs/artifacts
3. Expected outputs
4. Typical duration
5. Command to run

For implementation details (e.g., two-agent pattern, delegation mechanics), reference paw-workflow skill.

For "How do I start?", explain:

• PAW: New PAW Workflow command
• Parameters (branch, mode, strategy, issue URL)
• Available presets for quick setup (e.g., "use quick for minimal ceremony, thorough for full review pipeline")
• That prompt files generate on demand

Status Dashboard Format

Synthesize findings into sections:

• Artifacts: Existence and status
• Phases: Current progress (N of M)
• Phase Candidates: Pending/resolved candidate counts (if any exist)
• Branch & Git: Current state, divergence
• PRs: Open/merged status, review comments
• Next Actions: Recommended commands

Guardrails

• Verify phase count from ImplementationPlan.md—never assume
• Never mutate issue descriptions or PR content outside summary blocks
• Never push, merge, or rewrite git history
• Be idempotent: same state → same summary
• If required info missing, state blocker and resolution

Completion Response

Report:

• Current workflow state
• Recommended next action with command
• Any blockers or warnings
• PR/branch status summary