product-line

Pack Availability Guard

Before telling the user to run a skill from another project-local pack, check .agents/project.json.enabled_packs. If the target pack is not enabled, recommend /pack install <pack> instead of the target skill. Global skills are always valid. Skills from this same pack are valid because the current skill is already running from that pack.

Product Line — Product Path Portfolio Management

Manage product paths tracked in research/.progress.yaml. Product paths are research divergences — different products, apps, ICPs, expansion candidates, pivots, or competitive gaps — not git branches. Each path tracks its scope directory, status, evidence, pipeline stage, and revisit conditions.

Use activate when a deferred or revisit candidate path should become active research again. Use promote only when a research path graduates to an apps/{slug}/ code/app scope.

Prerequisites

research/.progress.yaml must exist for all modes except fork (which can create it). If missing, tell the user: "No product-path manifest found. Run /idea-scope-brief or /icp first — they create the manifest when multiple product directions emerge."

Manifest Schema

The canonical manifest lives at research/.progress.yaml:

active_paths:
  - "cost-intel"
max_concurrent: 1
product_paths:
  - id: "cost-intel"
    label: "Cost Intel"
    scope_path: "research/cost-intel/"
    status: active # active | deferred | archived | promoted | revisit_candidate
    source_skill: "icp"
    reason: "Why this path exists"
    archive_reason: null
    archived_at: null
    promoted_at: null
    evidence_refs: []
    revisit_trigger: null
    next_skill: "/competitive-analysis"
    pipeline_stage: "icp"
    last_touched: "YYYY-MM-DD"

Backward Compatibility

• Read legacy active_path as active_paths: [<value>] when active_paths is absent.
• Read legacy status: abandoned as status: archived.
• On writes, remove active_path, write plural active_paths, and prefer status: archived instead of abandoned.
• Treat status: promoted as graduated to code/app scope, not reactivated research.
• Exclude research/_archive/ paths from active-path selection and directory suggestions.

Modes

1. review — Portfolio Dashboard

Invocation: /product-line review

Present a dashboard comparing all paths:

Path	Status	Scope	Pipeline Stage	Last Touched	Evidence Maturity	Revisit Trigger

Group by active, deferred, revisit_candidate, archived, and promoted. For deferred and revisit_candidate paths, highlight the revisit trigger. For archived paths, show archive_reason and archived_at. For promoted paths, show promoted_at and the expected apps/{slug}/ scope.

Evidence maturity is derived from files under scope_path or evidence_refs:

• Early — only concept brief or initial ICP
• Developing — ICP plus competitive analysis, positioning, or journey map
• Mature — journey map plus metrics, GTM, monetization, or experiments
• Graduated — promoted to an app/code scope

When 3+ deferred or revisit-candidate paths exist with no recent activation, recommend /product-line triggers.

2. activate <path-id> — Reactivate Research

Invocation: /product-line activate <path-id>

1. Validate the path exists and has status deferred or revisit_candidate.
2. Check max_concurrent against the current active-path count. If at capacity, ask whether to deactivate another path or increase max_concurrent.
3. Ensure scope_path is not under research/_archive/. If it is archived, tell the user to run /product-line restore <path-id> first.
4. Scan existing research files to infer pipeline_stage.
5. Update the manifest:
   • Set status: active
   • Add the path ID to active_paths
   • Set last_touched to today
   • Set pipeline_stage from the scan when clearer than the existing value
6. Recommend the next skill from next_skill or the inferred pipeline stage.

3. archive <path-id> — Archive A Research Path

Invocation: /product-line archive <path-id>

1. Validate the path exists and is not already archived or legacy abandoned.
2. Ask for a concise archive rationale.
3. If scope_path is research/{slug}/ and the directory exists, create research/_archive/ as needed and move it to research/_archive/{slug}/.
4. Update the manifest:
   • Set status: archived
   • Remove the path ID from active_paths
   • Set scope_path to research/_archive/{slug}/ when the directory was moved
   • Set archive_reason, archived_at, and last_touched
5. Present confirmation with the archived directory and rationale.

4. restore <path-id> — Restore An Archived Path

Invocation: /product-line restore <path-id>

1. Validate the path exists and has status archived or legacy abandoned.
2. If scope_path is research/_archive/{slug}/ and the directory exists, move it back to research/{slug}/.
3. Ask whether to restore as deferred (default) or active. If active, enforce max_concurrent.
4. Update the manifest:
   • Set status: deferred by default, or active only when approved
   • Add to active_paths only when restored as active
   • Set scope_path back to research/{slug}/
   • Clear archived_at; keep the old archive rationale in reason or history, but set archive_reason: null
   • Set last_touched to today
5. Recommend /product-line activate <path-id> if restored as deferred.

5. promote <path-id> — Graduate To App Scope

Invocation: /product-line promote <path-id>

Use this only when research is ready to become or connect to an app under apps/{slug}/.

1. Validate the path exists and is not archived or legacy abandoned.
2. Derive {slug} from the path ID or scope_path.
3. Confirm or create apps/{slug}/ during the monorepo scaffolding flow. If scaffolding requires another skill, recommend /scaffold or the appropriate monorepo command instead of inventing app files here.
4. Keep research docs under research/{slug}/ as product evidence unless a separate scaffolder explicitly moves or links docs.
5. Update the manifest:
   • Set status: promoted
   • Remove the path ID from active_paths
   • Set promoted_at and last_touched to today
   • Keep scope_path pointing at the research evidence path
6. Present the app graduation result and the next implementation/scaffolding command.

6. fork <label> [--from <skill>] — Create A New Path

Invocation: /product-line fork "Enterprise vertical" --from competitive-analysis

1. Generate a stable product-path ID from the label.
2. Confirm that the slug is the intended product/app name, not merely an ICP, audience, or segment label.
3. Ask for supporting evidence, revisit trigger, and initial status (active or deferred).
4. Create or update research/.progress.yaml with the full schema fields.
5. Use scope_path: research/{slug}/.
6. If status is active, enforce max_concurrent and create research/{slug}/.
7. If status is deferred, create the manifest entry without forcing downstream research.

7. triggers — Check Revisit Conditions

Invocation: /product-line triggers

Scan paths with status: deferred or status: revisit_candidate and evaluate their revisit_trigger conditions:

• File-based triggers — check whether the referenced file exists.
• Evidence-based triggers — check active research files for relevant evidence.
• Time-based triggers — compare against today's date.
• External triggers — flag as requiring manual confirmation.

Classify each trigger as fired, pending, or manual. For fired triggers, recommend /product-line activate <path-id>.

Output

Update research/.progress.yaml. archive and restore may also move product-path directories between research/{slug}/ and research/_archive/{slug}/. promote may confirm or route creation of apps/{slug}/, but it does not move research docs.

Task Classification

When this skill produces follow-up work, file it by execution semantics:

• Immediately actionable implementation or documentation work goes in tasks/todo.md.
• Human-only external actions tied to automated steps go in tasks/manual-todo.md with _(blocks: Step N.X)_ or _(after: Step N.X)_; repo edits, SDK wiring, generated assets, local commands, tests, audits, and authenticated CLI/API work stay in tasks/todo.md.
• One-time condition-gated records, baselines, or future measurements go in tasks/record-todo.md with source, condition, non-blocking reason, evidence, and promotion rule.
• Cadence-based reviews, playtests, adoption checks, investor updates, retros, or docs-health checks go in tasks/recurring-todo.md with cadence, owner/agent, next due, evidence path, and escalation conditions.
• Do not put non-blocking records or recurring obligations in tasks/todo.md unless they have been explicitly promoted into current execution work.

Constraints

• Do not run downstream research or implementation skills.
• Do not treat archived, legacy abandoned, deferred, revisit-candidate, or promoted paths as active research targets.
• Do not use promote to mean "make active"; use activate.
• Do not use archive without a documented rationale.
• Respect max_concurrent for active paths.
• Use product-path terminology. Never use "branch" to mean a product path.

Alignment Page

When this skill produces durable deliverables (research, specs, plans, reports, prototypes, or any document output), build a full-depth HTML alignment page following ALIGNMENT-PAGE.md in this skill's directory. Output: alignment/product-line-{topic}.html.

Default Shipping Contract

Follow the shared shipping contract convention in CLAUDE.md.