legible-roadmap-with-sidequests

name	legible-roadmap-with-sidequests
description	Steward one canonical product roadmap while honoring an ADHD operator's energy-driven sidequests, using a link-or-opt-out discipline so every unit of work — planned or impromptu — stays traceable to real commits, PRs, or receipts. Use when designing how a roadmap coexists with burst-energy work, building a reconciliation cadence that folds sidequests back into the plan, or diagnosing a roadmap that has become either a rigid planning cage or an ignored wishlist. NOT for per-item issue-tracker mechanics (agent-issue-tracker-workflow), DAG decomposition of a single chosen problem (next-move), or PR description/authoring conventions (agent-pr-authoring).
license	Apache-2.0
allowed-tools	Read,Write,Edit,Bash,Grep,Glob
metadata	{"category":"Agent & Orchestration","tags":["roadmap-legibility","adhd-workflow","sidequest-capture","evidence-backed-progress","port-daddy"],"provenance":{"kind":"first-party","owners":["port-daddy"]},"pairs-with":[{"skill":"next-move","reason":"Once a sidequest or roadmap item is chosen and linked, hand it to next-move to decompose into a runnable DAG."},{"skill":"agent-work-receipt-designer","reason":"progressEvidence entries (commit/PR/receipt) should themselves be artifact-backed receipts, not self-reports."},{"skill":"agentic-coding-product-research","reason":"Recurring sidequest themes surfaced at reconciliation are candidate inputs to the next roadmap phase's research."}],"io-contract":{"kind":"deliverable","consumes":["[Truncated]"],"produces":["[Truncated]","[Truncated]"]}}

Legible Roadmap With Sidequests

Keep one product roadmap as the through-line while letting real, energy-driven sidequests happen — every unit of work traceable, nothing lost, nothing claimed without proof.

Use This For

• Designing a link-or-opt-out discipline so planned roadmap work and impromptu sidequests both stay traceable without a heavyweight planning tax on every impulse.
• Building the spawn-capture step that folds a sidequest's newly-surfaced follow-on work back into the roadmap instead of letting it evaporate.
• Running a periodic reconciliation pass that re-grounds burst-energy work against the current phases and catches drift early.
• Diagnosing whether a roadmap has become a rigid cage (sidequests get routed around it) or an ignored wishlist (status reported from optimism, not evidence).
• Auditing a work-unit log for the same shape of proof the roadmap-link CI gate demands of a PR: a real link, or an honest, explicit opt-out.

Do Not Use This For

• Per-item tracker mechanics like assignees, columns, or ticket lifecycle (agent-issue-tracker-workflow).
• Decomposing one already-chosen problem into a dependency-ordered DAG of subtasks (next-move).
• PR title/body/description conventions once a change is ready to land (agent-pr-authoring).

Process

flowchart TD
  A[Work happens: planned item or sidequest] --> B{Start gate}
  B -->|link| C[roadmapLink to existing item]
  B -->|opt-out| D[One-line optOutReason]
  C --> E[Work proceeds]
  D --> E
  E --> F{Did it spawn new durable work?}
  F -->|yes| G[Spawn-capture: create items or opt-outs for each]
  F -->|no| H[Attach progressEvidence at completion]
  G --> H
  H --> I[Periodic reconciliation pass]
  I --> J[Run scripts/roadmap_legibility.mjs]
  J -->|pass| K[Set next reconciliation date]
  J -->|fail| L[Fix work-unit records, re-run]

1. Confirm there is exactly one canonical roadmap. Two "sources of truth" make every downstream link ambiguous — consolidate or archive the rest before doing anything else.
2. At the start of any unit of work (planned or sidequest), pay the one-line cost: attach roadmapLink: <slug> if it obviously advances an existing item, or optOutReason: <one sentence> if it genuinely doesn't. Never route around this — a heavyweight gate at this step is what pushes momentum underground.
3. As work proceeds, capture real evidence — commit:<sha>, pr:<number>, receipt:<id> — as it happens, not retroactively from memory.
4. At completion, ask whether the work spawned new durable follow-on tasks. If yes, create roadmap items (or opt-outs) for each one before closing the loop — spawned work not captured in the same sitting tends to vanish.