roadmap
Skill14 starsChanged 5 days agoMITView on GitHub
Use when a project has to be declared above its pieces of work, from nothing or from the boards, plans and prototypes already made, when a slot has landed or the picture has changed and the roadmap must be redrawn, or when someone asks where the project stands. Produces and keeps the roadmap (iterations of deliverable slots, each slot's state derived from what exists for it) and the foundation reference (the technology decisions). Not for designing a slot (brainstorm), planning it (write-plan) or building it (execute-plan).
48 other projects use a version of this.
Adopted by
Most recently taken up by
- gopalparashar421/agent-skillstheir own version, first seen Nov 8, 2026
- dirien/yet-another-agent-harnesstheir own version, first seen Nov 8, 2026
- ryanmoelter/plumetheir own version, first seen Nov 4, 2026
- eyssette/js-templatetheir own version, first seen Nov 1, 2026
- danielcherubini/skillstheir own version, first seen Nov 1, 2026
- TheAxeC/ordotheir own version, first seen Nov 1, 2026
Used with
Projects with it also use sequence, stakeholder-update and 8 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.
GTMifySkillsequence
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
AnthropicSkillstakeholder-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
EpistatesSkillstrategy
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
ArvorCoSkillspecify
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
GitHubSkillprd
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
michalattSkillchallenge
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
Other versions48 projects use a version of this.
Word for word · 1
- Ovich/swarmstorm★ 14100% same
Edited · 0
None found.
Wrote their own version · 47
- cloudposse/atmos★ 1.4ksame name
- me2resh/apexyard★ 542same name
- andreaskelm/pm-brain★ 235same name
- genkovich/sdd★ 187same name
- YougLin-dev/Aha-Loop★ 181same name
- Ovid/paad★ 150same name
- facebookresearch/autoform-bot★ 136same name
- Mehdibargach/claude-code-pm-skills★ 97same name
- joshuadavidthomas/agent-skills★ 49same name
- spmtitsolution/OpenBA-Kit★ 38same name
37 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository34 other files it gives its agents.
- AGENTS.mdAGENTS.md
- brainstormSkill
- brainswarmSkill
- interviewSkill
- whiteboard-defenseSkill
- complexity-reworkSkill
- deepen-moduleSkill
- dependency-graphSkill
- simplify-reworkSkill
- simplifySkill
- write-diagramsSkill
- aiviewSkill
- swarm-boardSkill
- execute-planSkill
- long-horizonSkill
- orchestrate-planSkill
- guide-complexitySkill
- guide-createSkill
- guide-deepen-moduleSkill
- helpSkill
- next-stepSkill
- orchestrateSkill
- cold-readSkill
- write-planSkill
- technical-writingSkill
- write-conventionsSkill
- pr-reviewSkill
- recurring-review-findingsSkill
- swarmSkill
- design-prototypeSkill
- point-atSkill
- validate-deliverySkill
- verification-skill-createSkill
- create-workbenchSkill
The file
Files
Roadmap
The roadmap is the one document above the pieces of work: what is delivered in what order, what exists for each piece, what must be decided before the next one starts.
- A slot is something the customer can use or a system that can be deployed. Nothing smaller: a screen, a service, a migration, a board are pieces of work under a slot.
- A slot is done when the feature is delivered and working, never when its last plan finished.
- An iteration groups the slots delivered together.
- The first iteration of a project from scratch is usually one slot, the foundation: the smallest deployable system the features build on, reachable at the product's own domain, its first piece of work the repository itself, created at the provider the foundation names with its conventions file, before any code.
- No big design up front. The next iteration has its slots named precisely; later iterations carry a title, the outcome and at most a rough list. A slot is split into its pieces of work, or a new piece opened for a brainstorm, when the work reaches it.
The two documents
YYYY-MM-DD-<project>.roadmap.md, kindroadmap, dated at declaration and kept for the life of the project.foundation.reference.md, kindreference, undated, never retiring: the technology decisions, one row each; tags project androadmap.- Their contents and rows:
references/documents.md, read before drafting or redrawing either. - Both are registered and served via the
aiviewskill in the product's group, the one its product-level boards and prototypes share, created and titled after the product when none exists, the roadmap started when the declaration began. - Documents in the project's folder that are not the product's are linked apart, as not its slots.
Slots and their state
- Every document of a slot carries the slot's slug as a tag, whatever group it sits in.
- The state is derived from what the tag finds, never typed: empty (nothing), in design (a board, or a plan not yet approved), drawn (prototypes), planned (an approved plan), in progress (a plan's tracker has a ▶), landed (the person confirmed the feature delivered and working, with the evidence: a URL, a deployment, a customer).
- In the tracker only in progress is ▶ and only landed is ✅; in design, drawn and planned stay ⬜ with the state written in the slot row.
- At declaration the draft derives each state from the reading, and the tags are applied when the interview confirms which slot a document serves.
- A board serves a slot, or the roadmap itself when the idea is product-level, and its tag says which.
- The roadmap resolves its own decisions through the
interviewskill, never through a board of its own, and keeps the exchange in its interview log: the question with its options and recommendation, then the answer as given.
Declare
- Read everything that exists for the project: its documents in aiview, its code and conventions file, any register the person names.
- Name what is missing: slots no document covers (accounts and login, billing, the foundation itself are the usual absences), foundation rows no decision covers, dependencies between slots that nothing states.
- Draft both documents whole, a recommendation in every open place with its reason, register and serve them, tell the person the URLs, and hand them over as something to correct.
- Interview on the draft. The tree: the order of iterations, which slot each existing piece belongs to, the foundation rows the first slot needs. Rows a later slot needs stay open with that slot named. Every answer lands in the documents as it is given. For the name and domain rows, ask the person to check the candidate domains at their registrar; a name whose domain is taken is never the recommended one.
- Close. The state node names the first slot and its next step. Say so in chat.
Redraw
Run when a slot lands, a plan finishes, a board opens for a slot, or the person says the picture changed.
- Derive every state again. Mark landed only with the person's evidence.
- Split the slot the work has reached into its pieces of work, each a group.
- Ask the foundation rows the next slot needs and refuse to start it while one is open.
- Deviate from the earlier drawing when the last slot taught something, and write what changed and why in the redraw log.
- Retire the landed slot's groups as the aiview contract asks.
Status
Read-only, in chat: the iteration, the slot in progress, what blocks, the foundation rows still open for the slot ahead, the pieces of work the next slot still lacks.
The diagram
The roadmap opens on its tracker, drawn to the tracker protocol of the write-plan skill (its references/tracker.md, the Slots section): one subgraph per iteration, one node per slot with the glyph of its derived state, dependency edges, the state node with the iteration, the slot in progress, next, blocked and the open rows. The coarse iterations share one subgraph, "later", their order carried by the edges alone. Look at it in aiview after every edit: a block that fails to parse shows mermaid failed.
Red flags
| Thought | Reality |
| "The stack can be decided when the first plan is written" | Then it is decided in one plan, and the second slot inherits it without seeing it. Rows close in the foundation, before the slot. |
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
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.

