roadmap
Skill1 starsChanged 57 days agoApache-2.0View on GitHub
Find the next thing to work on in wilanis-js and take it from issue to pull request. Use when asked "what should I work on", "pick a task", "take issue N", or to start work on a roadmap item.
- Deletes or force-pushes
- Commits and pushes
51 other projects use a version of this.
Adopted by
Most recently taken up by
- liuhlab/liulab-skillstheir own version, first seen Nov 16, 2026
- lengrufeng2008/dsh-skill-librarytheir own version, first seen Nov 15, 2026
- GravionLabs/skillstheir own version, first seen Nov 15, 2026
- Palatine-Labs/conisotheir own version, first seen Nov 15, 2026
- iuliandita/skillstheir own version, first seen Nov 15, 2026
- yarikleto/claude-teacher-plugintheir own version, first seen Nov 14, 2026
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.
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 versions51 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 51
- 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★ 154same name
- facebookresearch/autoform-bot★ 143same name
- Mehdibargach/claude-code-pm-skills★ 97same name
- joshuadavidthomas/agent-skills★ 49same name
- spmtitsolution/OpenBA-Kit★ 38same name
41 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 repository4 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- coordinateSkill
- reviewSkill
- rfcSkill
The file
We split this skill into the rules it gives an agent
7 rules. 0 are common, 3 are distinctive, and 4 are very specific to this project.
0
Common
Most of the 44 edited versions we read keep these.
3
Distinctive
This author's own choices. Worth a look.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 58
Only once the push is yours does the rest follow -- these record the claim, they do not make it, so a failure here is worth retrying and …
- Very specific: line 67
If --add-assignee shows someone else already there, they claimed without a branch.
- Distinctive: line 70
Before taking anything, check both: the label and the ref.
- Distinctive: line 80
Do only the step. If the step needs something the RFC does not say, stop and ask.
- Very specific: line 81
Every new checker rule takes the next free code in its family, the hint the RFC wrote, and a sabotage test in …
- Very specific: line 90
Commit in plain imperative sentences (no trailers).
- Very specific: line 123
If the demo docs/roadmap.md promises for the milestone does not run by hand, the milestone is not closed, whatever its issues say.
“Common” is measured: kept by at least 6 in 10 of the 44 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.
Files
Working the roadmap
The roadmap is docs/roadmap.md: demos, each drawing on RFCs under docs/rfcs/, tracked as issues on
wilanis/wilanis-js under the milestone of the demo that first shows them. Read CONTRIBUTING.md once;
this skill is the procedure.
1. Find
One milestone is worked at a time: the open one with the earliest due date. docs/roadmap.md numbers the
milestones in the order their RFCs allow, so the current one is the lowest open number and its date says so.
gh api repos/wilanis/wilanis-js/milestones --jq '[.[] | select(.state == "open")] | sort_by(.due_on)[0].title'
gh issue list -R wilanis/wilanis-js --label status:ready --search "no:assignee" \
--milestone "<that title>" --json number,title,labels,milestoneWithin the milestone, prefer the task whose RFC's earlier steps are closed. If nothing there is
status:ready, its remaining steps wait on one in flight: say which, and do not reach into the next
milestone. If the user named an issue, use it. Never take an issue whose RFC is status:draft: say so and
stop.
2. Read
- The issue: which RFC and which step of its Implementation plan.
- The RFC file, whole: the Reference section states every rule, code, hint and test the step needs.
CLAUDE.md, the "How to change things" entry that matches the step (a new rule, a new kind, ...).
3. Claim, by pushing the branch
The claim is the branch on the remote, not the assignee. Assigning is additive: GitHub answers the same
whether or not someone else is already assigned, so two authors who read no:assignee a second apart both
"claim" and both start. Creating a ref is the one operation the server refuses twice. So the branch is
pushed first, empty, before a line is written, and whoever loses that race stops.
git switch main && git pull --ff-only
git switch -c N-short-title # the issue number first: the branch-name check requires it
git push --force-with-lease=refs/heads/N-short-title: -u origin N-short-titleThe lease is empty on purpose: it says this ref must not exist. If it does, the push is rejected
(stale info) and the task is already taken. Do not force, do not pick another name for the same issue,
do not carry on locally. Say who holds it and pick something else:
gh pr list -R wilanis/wilanis-js --head N-short-title --state all --json author,urlA plain git push -u origin N-short-title is not enough: if the other author has pushed a commit, a
fast-forward would hand you their branch and both of you would work on it.
Only once the push is yours does the rest follow -- these record the claim, they do not make it, so a failure here is worth retrying and never worth abandoning the branch over:
gh issue edit N -R wilanis/wilanis-js --add-assignee @me
gh project item-edit 1 --owner wilanis --field Status --value "In Progress" \
--url https://github.com/wilanis/wilanis-js/issues/NIf --add-assignee shows someone else already there, they claimed without a branch. Stop and ask; the
branch says who is working, the assignee says who said they would.
Before taking anything, check both: the label and the ref.
gh issue list -R wilanis/wilanis-js --label status:ready --search "no:assignee" \
--milestone "<that title>" --json number,title
git ls-remote --heads origin 'refs/heads/N-*' # empty means nobody holds it4. Implement
Do only the step. If the step needs something the RFC does not say, stop and ask; do not widen it.
Every new checker rule takes the next free code in its family, the hint the RFC wrote, and a sabotage
test in packages/runtime/test/example.test.ts (or the plugin's test directory). Run:
npm test5. Deliver
Commit in plain imperative sentences (no trailers). Push and open the pull request from
.github/PULL_REQUEST_TEMPLATE.md, with Closes #N in the body:
gh pr create -R wilanis/wilanis-js --fill --body "Closes #N"Tick the step in the tracking issue's checklist once merged. If it was the last of its RFC, edit the
RFC's status to implemented and the row in docs/rfcs/README.md in the same pull request, and close
the tracking issue when that merges. The state of an RFC is its status:* label and its header, kept
equal; the board shows them and adds nothing.
Label the RFC's next step status:ready, or the board stops answering what can be taken and the milestone
looks blocked when it is not.
Merging deletes the branch on the remote and not the copy here. Delete that too, or the clone collects branches whose remote is gone:
git switch main && git pull --ff-only && git branch -D N-short-title
git fetch --prune6. When the milestone closes
Close the GitHub milestone, then point the board's Now view at the next one, or it keeps showing a milestone that is done:
gh api graphql -f query='mutation($v:ID!,$f:String!){ updateProjectV2View(input:{viewId:$v,filter:$f}){ projectV2View { name filter } } }' \
-f v=PVTV_lADOE33gE84Bi9UDzgLogF0 -f f='milestone:"<the next milestone title>"'If the demo docs/roadmap.md promises for the milestone does not run by hand, the milestone is not closed,
whatever its issues say.
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.

