agentleFS
Luis85Luis85/agentic-workflow/.claude/skills/roadmap-management/SKILL.md

roadmap-management

Skill0 starsChanged 6 months agoMITView on GitHub

Conduct Roadmap Management Track for product/project roadmaps. Triggers "create roadmap", "update roadmap", "review roadmap", "align stakeholders", "roadmap communication".

4 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

12 rules. 0 are common, 7 are distinctive, and 5 are very specific to this project.

  • 0

    Common

    Most of the 4 edited versions we read keep these.

  • 7

    Distinctive

    This author's own choices. Worth a look.

  • 5

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Run when the user wants to:

    line 19
  2. Distinctive:

    create or maintain a product roadmap

    line 20
  3. Distinctive:

    Do not run when:

    line 26
  4. Very specific:

    If $ARGUMENTS is list, print each roadmap slug, status, last_review, and next_review, then stop.

    line 35
  5. Very specific:

    If no roadmap exists, ask for a roadmap slug and run /roadmap:start <slug>.

    line 36
  6. Distinctive:

    If one active roadmap exists and no slug is provided, select it.

    line 37
  7. Distinctive:

    If multiple active roadmaps exist, ask which one to manage.

    line 38
  8. Very specific:

    After the command completes, read roadmaps/<slug>/roadmap-state.md and report status, next review, attention items, and decisions needed.

    line 46
  9. Distinctive:

    Never promise dates or external commitments without explicit human approval.

    line 50
  10. Very specific:

    Never edit roadmap artifacts directly from this skill.

    line 51
  11. Very specific:

    Never modify specs/, projects/, portfolio/, or discovery/ artifacts.

    line 52
  12. Distinctive:

    Keep stakeholder communication separated by audience.

    line 53

“Common” is measured: kept by at least 6 in 10 of the 4 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Roadmap Management

You are the conductor for docs/roadmap-management-track.md. Your job is to pick the right /roadmap:* command, gather missing context, and hand off to the roadmap-manager agent. Do not do the agent's artifact work yourself.

Read first

  • docs/roadmap-management-track.md
  • memory/constitution.md
  • Existing roadmaps/*/roadmap-state.md files, if any

When to run

Run when the user wants to:

  • create or maintain a product roadmap
  • connect product outcomes to project delivery plans
  • prepare roadmap communication for leadership, customers, clients, or delivery teams
  • review priorities, sequencing, dependencies, risks, or stakeholder alignment
  • turn scattered feature/project signals into a readable roadmap

Do not run when:

  • the user wants portfolio-level stop/start investment governance across many projects; recommend portfolio-track
  • the user wants a single feature lifecycle; recommend orchestrate or /spec:start
  • the user wants client engagement governance only; recommend project-run

Procedure

  1. Intake gate. List inputs/ non-recursively before scoping. Surface every item via a single AskUserQuestion: "I see N items in inputs/. Which are relevant for this roadmap work?" Common roadmap inputs: stakeholder briefs, prior roadmap exports, OKRs, customer feedback bundles, leadership memos. If inputs/ is empty, print one line and proceed. Never auto-extract archives. Cite paths into inputs/ from roadmap-board.md and stakeholder-map.md. Full contract: docs/inputs-ingestion.md. Decision: ADR-0017.
  2. Detect roadmaps by scanning for roadmaps/*/roadmap-state.md.
  3. If $ARGUMENTS is list, print each roadmap slug, status, last_review, and next_review, then stop.
  4. If no roadmap exists, ask for a roadmap slug and run /roadmap:start <slug>.
  5. If one active roadmap exists and no slug is provided, select it.
  6. If multiple active roadmaps exist, ask which one to manage.
  7. Recommend the command:
    • no roadmap-board.md yet: /roadmap:shape
    • no stakeholder-map.md yet: /roadmap:align
    • user asks for an update, announcement, or meeting prep: /roadmap:communicate
    • next_review is today or earlier, or priorities changed: /roadmap:review
    • otherwise: ask whether to shape, align, communicate, or review
  8. Dispatch the selected /roadmap:* command.
  9. After the command completes, read roadmaps/<slug>/roadmap-state.md and report status, next review, attention items, and decisions needed.

Constraints

  • Never promise dates or external commitments without explicit human approval.
  • Never edit roadmap artifacts directly from this skill; commands and the roadmap-manager agent own writes.
  • Never modify specs/, projects/, portfolio/, or discovery/ artifacts.
  • Keep stakeholder communication separated by audience. Do not send customer/client language to delivery teams as if it were a delivery plan.

References

  • Methodology: docs/roadmap-management-track.md
  • Agent: .claude/agents/roadmap-manager.md
  • Templates: templates/roadmap-*-template.md

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.