agentleFS
ryanmoelterryanmoelter/plume/.claude/skills/roadmap/SKILL.md

roadmap

Skill0 starsChanged 10 days agoApache-2.0View on GitHub

Read and change Plume's roadmap, which lives in Linear. Use for "what should I work on", "what's next", "what's in the queue", to record that an item shipped, to add an item, or to reprioritize the queue.

48 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use sequence, stakeholder-update and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • GTMifySkill

    sequence

    Build a multi-channel outreach cadence (email + LinkedIn + phone) for a target persona or named prospect. Use when the user says 'build a sequence', 'outreach cadence', 'multi-touch sequence', 'multi-channel cadence', 'follow-up sequence', or needs a coordinated multi-channel play. Do NOT use for single cold emails — use cold-email instead.

    Also in 3 projects that use this

  • AnthropicSkill

    stakeholder-update

    Generate a stakeholder update tailored to audience and cadence. Use when writing a weekly or monthly status for leadership, announcing a launch, escalating a risk or blocker, or translating the same progress into exec-brief, engineering-detail, or customer-facing versions.

    Also in 3 projects that use this

  • EpistatesSkill

    strategy

    Plan weekly or monthly X content strategy. Use when the user wants to plan their content calendar, set growth goals, build a content pipeline, or think strategically about their X presence.

    Also in 3 projects that use this

  • ArvorCoSkill

    specify

    Create feature specification from natural language description. Use when starting a new feature or creating a spec. Triggers on: create spec, specify feature, new feature spec.

    Also in 3 projects that use this

  • GitHubSkill

    prd

    Generate high-quality Product Requirements Documents (PRDs) for software systems and AI-powered features. Includes executive summaries, user stories, technical specifications, and risk analysis.

    Also in 7 projects that use this

  • michalattSkill

    challenge

    Pressure-test plans, decisions, and directions by picking and answering the most relevant challenge questions for the current conversation

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

5 rules. 0 are common, 2 are distinctive, and 3 are very specific to this project.

  • 0

    Common

    Most of the 41 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 3

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    When an issue could fit two projects:

    line 25
  2. Very specific:

    Call list_issues with team: "Plume", state: "unstarted", and fields: ["id", "title", "priority", "estimate", "project", "description"].

    line 51
  3. Distinctive:

    Do not write a summary of what shipped into the issue.

    line 61
  4. Very specific:

    If the work only partly landed, leave the issue open and add a save_comment saying what is left.

    line 63
  5. Very specific:

    Do not inventory the codebase in the description.

    line 69

“Common” is measured: kept by at least 6 in 10 of the 41 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

The roadmap is Linear, team Plume. Projects are organized by use case. Route an issue by its primary user outcome, not by the UI component or code area it touches:

  • Multitask — Home, the sidebar, agent status, notifications, unread state, tabs and splits, terminal history, worktrees and directories, workstream coloring, cross-conversation search.
  • Core chat — the composer, message rendering, selection and copy, conversation controls, history and transcript continuity, remote-session reconnection, model and effort controls, Claude Code mods, cross-session messaging.
  • Understand context — the info pane, reference chips, session metadata, quotas, statusline customization, goal status, issue/PR/MR references, live app previews, the /mcp flow.
  • Review — PR visibility, checks, stacks, diffs, reviewing agent output, review-specific visualization.
  • Tools for agents — user-facing plume commands and capabilities agents invoke, including bundled Plume skills.
  • Test on devices — Android devices and emulators, iOS Simulator, remote devices, embedded device views, streaming, device-specific split workflows.
  • Get started — setup, permissions, onboarding, imports and migration.
  • Codex support — Codex-specific compatibility and missing capabilities.
  • Delight — palettes and visual polish that make Plume feel better to use.
  • Infrastructure — shared engineering foundations, process and socket plumbing, testing, window-resize performance, titlebar theming.
  • Integrations — authentication, external-provider support, sync, connector configuration.
  • Admin — releases, documentation, operational and non-code work.
  • Code architecture visualizer — its own project.
  • Later — explicitly deferred work, including driving other apps through a pluggable desktop-control backend.

When an issue could fit two projects:

  • Device work goes in Test on devices, even when an agent invokes it through the CLI.
  • Codex support holds only Codex-specific gaps. A general improvement found through Codex goes in its outcome project.
  • User-facing reference chips go in Understand context, not Integrations.
  • The command an agent invokes goes in Tools for agents. The shared plumbing under it goes in Infrastructure.

Six projects are retired and Completed: Agent & session, Terminals, tabs & window, Sidebar & tasks, Composer & statusline, Chat rendering & interaction, and Requests. Never put an issue in one. Older Done history still lives there, so a search can still return them.

To do means queued. Backlog means wanted but not queued. In Progress, Done and Canceled mean what they say.

Status names get renamed, so filter on the state type when reading: unstarted is To do, backlog is Backlog. A name that matches nothing returns an empty list, not an error.

estimate is the size: 1 = XS, 2 = S, 3 = M, 5 = L, 8 = XL. Unset means nobody has sized it. Those are the only values the scale accepts — anything else snaps to the nearest.

Labels:

  • Requested means someone outside this repo asked for the item. It records provenance only, so choose the project by outcome as usual.
  • Needs decision means the decision in the description must be settled first.
  • Bug marks a defect.
  • The Project snapshot · 2026-11-02 group records each issue's project before the migration. Leave it on existing issues, and never apply it to a new one.

An issue states the item and any real blocker. It does not describe what the code already does — read the code for that.

What should I work on

Call list_issues with team: "Plume", state: "unstarted", and fields: ["id", "title", "priority", "estimate", "project", "description"].

Sort by priority ascending. Priority is the queue order.

Report a numbered list: identifier, title, size, project. Read the whole description before starting anything.

Record that an item shipped

save_issue with the identifier and state: "Done". Nothing else.

Do not write a summary of what shipped into the issue. The commit is the record — this is what stops Linear regrowing the history the roadmap accumulated.

If the work only partly landed, leave the issue open and add a save_comment saying what is left.

Add an item

save_issue with team: "Plume", the project for its outcome, state: "Backlog", and a description holding the item and any real blocker. If the item came from outside this repo, add the Requested label.

Do not inventory the codebase in the description. Ask the user for a size; leave estimate unset rather than guessing.

Reprioritize

Into the queue: state: "To do" plus a priority. Out of it: state: "Backlog" and priority: 0.

Reordering means changing priority. Never record an order in the description — priority already holds it.

Present it back

Group by project and name the identifier, so the reader can open it. Say the title and the size; offer the detail rather than pasting a whole description into chat.

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.