triage
Skill1 starsChanged 8 days agoMITView on GitHub
Use when you start any task in a repository that uses dkstack. Selects the guidelines and skills that apply to the task.
189 other projects use a version of this.
Adopted by
Most recently taken up by
- udecode/platetheir own version, first seen Nov 8, 2026
- vercel-labs/add-skilltheir own version, first seen Nov 8, 2026
- chrysogonus/power-agentstheir own version, first seen Nov 8, 2026
- ishandutta2007/Awesome-Claude-Skillstheir own version, first seen Nov 8, 2026
- Adelie-Squad/solosquadtheir own version, first seen Nov 8, 2026
- cloudflare/vinexttheir own version, first seen Nov 8, 2026
Used with
Projects with it also use writing-beats, setup-matt-pocock-skills 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.
LesaniSkillwriting-beats
Shape an article as a journey of beats, choose-your-own-adventure style. The user picks a starting beat from the raw material, you write only that beat, then offer options for where to pivot next, beat by beat, until the article reaches a natural end. Use when the user has raw material and wants to assemble it as a narrative rather than an argument.
Also in 66 projects that use this
GoogleSkillsetup-matt-pocock-skills
Configure this repo for the engineering skills — set up its issue tracker, triage label vocabulary, and domain doc layout. Run once before first use of the other engineering skills.
Also in 167 projects that use this
liufeng1976Skillsetup-ts-deep-modules
Wire dependency-cruiser into a TypeScript repo so each package is a deep module — implementation hidden in subfolders, reachable only through its entry-point files. User-invoked.
Also in 42 projects that use this
SethGammonSkillascii-diagram
Generate perfectly aligned ASCII diagrams — architecture, flow, sequence, box-and-arrow. Uses a programmatic character-grid approach so alignment is guaranteed by math, not token prediction. Includes post-render verification.
Also in 4 projects that use this
simo97Skillmigrate-to-shoehorn
Migrate test files from `as` type assertions to @total-typescript/shoehorn. Use when user mentions shoehorn, wants to replace `as` in tests, or needs partial test data.
Also in 68 projects that use this
LesaniSkillwriting-shape
Take a markdown file of raw material and shape it into an article through a conversational session — drafting candidate openings, growing the piece paragraph by paragraph, arguing about format (lists, tables, callouts, quotes) at each step. Use when the user has a pile of notes, fragments, or a rough draft and wants help turning it into something publishable.
Also in 63 projects that use this
- +1 more from Lesani
Other versions189 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 189
- mattpocock/skills★ 309ksame name
- withastro/astro★ 64ksame name
- vectorize-io/hindsight★ 60ksame name
- ogulcancelik/herdr★ 47ksame name
- herdrdev/herdr★ 47ksame name
- colinhacks/zod★ 44ksame name
- QuivrHQ/quivr★ 40ksame name
- The-Vibe-Company/quivr★ 40ksame name
- vercel-labs/add-skill★ 39ksame name
- vercel-labs/skills★ 39ksame name
179 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 repository12 other files it gives its agents.
- AGENTS.mdAGENTS.md
- apply-architectureSkill
- bootstrapSkill
- clarify-intentSkill
- code-reviewSkill
- map-workSkill
- plan-workSkill
- releaseSkill
- tdd-loopSkill
- write-adrSkill
- write-prSkill
- writing-checkSkill
The file
Show its rules and text
We split this skill into the rules it gives an agent
13 rules. 0 are common, 10 are distinctive, and 3 are very specific to this project.
0
Common
Most of the 176 edited versions we read keep these.
10
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
- Very specific: line 8
Use this skill first when you work in any repository that points to dkstack.
- Distinctive: line 14
If a feature plan exists, follow feature-plans.md.
- Distinctive: line 18
Read the section index or skill, then the detailed files it links to.
- Distinctive: line 20
If the task spans multiple areas, read each relevant item.
- Distinctive: line 33
If you cannot start subagents, do the steps one at a time in the order of this skill.
- Distinctive: line 44
Use it when you write or materially update the body.
- Distinctive: line 54
Read the issue before you select a skill.
- Distinctive: line 108
Read it when a human asks for the analysis, or when the task has one of these signals:
- Distinctive: line 118
Do not add a phase, a checklist or a report for ordinary work.
- Distinctive: line 122
Do not read every file. Read only what the task needs.
- Very specific: line 123
If two dkstack rules conflict, follow the more specific rule.
- Distinctive: line 125
If no section clearly applies, read architecture/index.md and coding/index.md.
- Very specific: line 126
After you finish, dkstack remains the source of truth.
“Common” is measured: kept by at least 6 in 10 of the 176 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
Triage
Use this skill first when you work in any repository that points to dkstack.
Decide which sections and skills apply to the current task. Read only those.
Steps
If a feature plan exists, follow feature-plans.md.
- Identify the type of work: design, implementation, documentation, version control, review, workflow, or mixed.
- Select the relevant sections and skills from the lists below.
- Read the section index or skill, then the detailed files it links to.
- Apply the rules. Prefer the smallest set that covers the task.
- If the task spans multiple areas, read each relevant item.
Subagents
If you can start subagents, use them for step 3 when the task spans more than one area. Do steps 1 and 2 yourself. Then read the selected sections and skills at the same time, each in its own subagent. Give each subagent a brief that is complete without your context:
- The task, and the type of work from step 1.
- One section or skill from step 2 to read, with its link.
- What to report: the rules that apply to the task, each with its link, and a word limit, for example 300 words.
- This line: "Report to me only. Do not change files, commit, push, post, approve or merge."
Then do steps 4 and 5 yourself. A report from a subagent is a claim. Check each rule that you apply at its source.
If you cannot start subagents, do the steps one at a time in the order of this skill.
Skills
- New or non-compliant repository: bootstrap
- Non-trivial change: clarify-intent
- Design or structure: apply-architecture
- Plan or resume a feature: plan-work
- Map a multi-PR objective: map-work
- Implementation: tdd-loop
- Pull request body: write-pr. Use it when you write or materially update the body. Do not use it for a review that leaves the body unchanged.
- Review: code-review
- Bug or unexpected behavior: coding/debugging.md
- Durable decision: write-adr
- Any text: writing-check
- Version number or release: release
Work route
Read the issue before you select a skill. A missing lifecycle or eligibility does not stop triage. An authorized task carries permission through PR creation and review fixes. Repair unambiguous missing metadata without asking permission again; never override a human-only or conflicting designation. Classify the request. Verify chat instructions under owner instructions in chat. For an explicit instruction from the verified repository owner to implement a named issue, follow the owner-directed transition in work.md: fill missing non-conflicting tracker metadata, establish readiness, re-read tracker state and claim before implementation. Do not infer approval from issue authorship or a request to triage. Do not request repeated approvals for in-scope steps after an authenticated implementation instruction. Check implementation readiness in work.md before implementation. Run preflight before a tracker write. One pull request uses plan-work. More than one pull request uses map-work, then that session stops.
Scale
The ladder is a guide, not a gate. Use the smallest set of skills. A small task uses one or two skills. It does not go through every step.
One pull request, whether small or medium, uses plan-work. A large objective can also use plan-work when it fits one coherent pull request. More than one pull request uses map-work to define the issue graph before any child implementation. Do not create children only because a task has many commits or files.
- Bug fix: tdd-loop, then code-review.
- Feature: start with clarify-intent and apply-architecture. For a multi-PR objective, use map-work, then stop that session. For one-PR work, use plan-work, tdd-loop, then code-review. Use writing-check when the change has text that people or agents read. Use release only when the task is a release.
- Release: release, after the change is reviewed.
Handoffs:
- A review finding goes back to the implementation step.
- A design finding goes back to the architecture step.
One agent can do all the steps. Many agents can share the steps. A skill does not make one role into one agent.
Section map
- Design or structure changes: architecture/index.md
- Code changes: coding/index.md and tdd/index.md
- Commits or branches: version-control/index.md
- Version numbers or releases: versioning/index.md
- Documentation, comments, or messages: writing/index.md
- Any text: writing/index.md
- Issues, pull requests, review, merge or release: workflow/index.md
Stewardship
Routine work does not read stewardship. Read it when a human asks for the analysis, or when the task has one of these signals:
- A new public interface.
- A changed module boundary or a new dependency.
- A workaround that avoids the current architecture.
- The same knowledge appears again.
- Two principles conflict.
- A trade-off affects compatibility, performance or security.
- An ADR no longer matches the code or the current decision.
Do not add a phase, a checklist or a report for ordinary work. The page holds the rules.
Rules
- Do not read every file. Read only what the task needs.
- If two dkstack rules conflict, follow the more specific rule. If a project rule conflicts with a dkstack rule, follow the precedence rules in AGENTS.md.
- To decide between two designs that both work, use the easier to change rule in architecture/index.md.
- If no section clearly applies, read architecture/index.md and coding/index.md.
- After you finish, dkstack remains the source of truth.
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.

