Agent skill

Implement

by RafaelGB in RafaelGB/Obsidian-ZettelFlow

Stage 4 of the SDD pipeline — execute a ZettelFlow task checklist (in a GitHub issue comment) test-first (red → green → refactor), one commit per advance, keeping npm run verify green and CI green…

MITAuto-check passedTesting & QA

Install Implement

skills CLI
$ npx skills add RafaelGB/Obsidian-ZettelFlow --skill implement -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install RafaelGB/Obsidian-ZettelFlow implement --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/RafaelGB/Obsidian-ZettelFlow.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.claude/skills/implement .claude/skills/implement && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
implement
GitHub stars
174
Token cost
~1.7k tokens
SKILL.md length
952 words
Files
1
Skills in repo
9
Repo updated
First seen
Licence
MIT

At a glance

Stage 4 of the SDD pipeline — execute a ZettelFlow task checklist (in a GitHub issue comment) test-first (red → green → refactor), one commit per advance, keeping npm run verify green and CI green…

  • Works in 5 steps: Docs audit (mandatory) → README placement audit — by door rank… → Walk the verification script (mandatory) → …
  • Tasks that involve Test-driven development
  • SKILL.md covers Before starting, The loop (per task), Rules and Exit → stage 5
  • Calls npm, gh and git

What it does

Implement is an agent skill from RafaelGB/Obsidian-ZettelFlow. Stage 4 of the SDD pipeline — execute a ZettelFlow task checklist (in a GitHub issue comment) test-first (red → green → refactor), one commit per advance, keeping npm run verify green and CI green on every commit on a single feature branch. Use after a tasks comment exists and the user says "implement", "build issue

Its SKILL.md is about 1.7k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Testing & QA, covering Test-driven development. It works with npm, GitHub and Obsidian. The repository describes itself as: ZettelFlow is a plugin for Obsidian that helps you to create and manage your notes in a Zettelkasten way. The licence is MIT.

When your agent uses it

  • Tasks that involve Test-driven development

Example prompts

  • “implement”
  • “/implement”

Requirements

  • Node.js

Workflow steps

5 steps, taken from the step headings in SKILL.md.

  1. Docs audit (mandatory)
  2. README placement audit — by door rank (mandatory for user-facing features)
  3. Walk the verification script (mandatory)
  4. Quality check
  5. Close the issue via PR

What it can do on your machine

Read from SKILL.md and the folder at commit 0676350. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • npm
    • gh
    • git
    • npx

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md. Its commands use npm, gh, git and npx, which can reach the network depending on how they are called.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Implement loads about 1.7k tokens when it runs. Until then it costs about 82 tokens; SKILL.md has 952 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~82
When it runs · the whole SKILL.md, loaded when a task matches
~1.7k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from RafaelGB/Obsidian-ZettelFlow at commit 0676350, republished under its MIT licence (© RafaelGB). 952 words, ~1,737 tokens.

Download SKILL.mdSave it as .claude/skills/implement/SKILL.md (or your agent's skills folder).
name
implement
description
Stage 4 of the SDD pipeline — execute a ZettelFlow task checklist (in a GitHub issue comment) test-first (red → green → refactor), one commit per advance, keeping npm run verify green and CI green on every commit on a single feature branch. Use after a tasks comment exists and the user says "implement", "build issue

/implement — build the tasks test-first

Stage 4 of the SDD pipeline. Work the task checklist (in the GitHub issue comment) top to bottom using the tdd discipline. This stage is done by the main assistant (not a subagent) because it commits to the branch and must keep CI green at every step.

Before starting

Run gh issue view <N> — read the spec (body) and the plan + tasks comments. The tasks comment has the checklist; the plan comment has the files, guardrails, and risks.

The loop (per task)

  1. Red — write/extend the test named in the task; run npm test and watch it fail for the right reason. Import from @jest/globals; resolve source via the bare aliases; extend test/__mocks__/obsidian.ts when the unit pulls in more Obsidian API (see the tdd skill).
  2. Green — make the minimal change to pass.
  3. Refactor — clean up with the suite green.
  4. Verify — npm run verify (typecheck + oxlint + jest) must be green. For score tasks also run npm run lint:obsidian and confirm no new violations. For i18n tasks confirm en.ts/es.ts key parity.
  5. Commit — one Conventional Commit per completed task/advance (constitution §IX): git add -A && git commit -m "<type>(<scope>): <subject>".
  6. Check off the task — mark [x] directly in the GitHub issue comment (edit the comment with gh issue comment <comment-id> --edit-last or find the comment id and patch it).
  7. Keep CI green — push the branch; the CI workflow re-runs the blocking guardrails. Because verify is green locally, CI stays green. Fix forward if a push ever goes red — don't stack more commits on a red branch.

Rules

  • Single branch — one feature/* branch for the whole issue; never commit to main.
  • Never introduce innerHTML, inline el.style.*, Title-case UI strings, bare console.*, or global app — they cost score (constitution §III–IV). Build DOM with createEl; style with c() + SCSS; log with log.
  • The user's theme wins (§XV). No hex or named colour in a stylesheet, no pixel the --size-4-* grid can express, Obsidian's own classes (mod-cta, clickable-icon, setting-item, is-active) before inventing one, and a shape that exists twice goes in src/styles/utils/mixins.scss. The guide, with Obsidian's own wording, is docs/development/obsidian-styling.md.
  • Touching the Canvas patcher? Keep every patched access guarded and uninstalled on unload (§VI) — see issue #91 / the reviewer agent.
  • Update the matching docs/ page + mkdocs.yml nav in the same commit that changes behavior (§VIII).

Exit → stage 5

When all tasks are checked off, do four things before declaring done:

1. Docs audit (mandatory)

For every user-facing change — new feature, changed behaviour, new config option, new action, changed UI text — ask: does the existing docs page cover this?

  • New feature / behaviour change → update the matching page under docs/ AND check the mkdocs.yml nav (add an entry if the feature deserves its own page).
  • New action → add or update docs/actions/<ActionName>.md.
  • New config option → update the settings section of the relevant architecture page.
  • API / public surface change → update docs/api/ZettelFlowAPI.md.
  • No user-facing change (pure refactor, test, chore) → docs audit is still required; confirm explicitly that no doc update is needed and state why.

This audit is a blocking exit criterion — do not commit the implementation without it. Docs and code travel in the same commit (or a docs: follow-up commit immediately after).

Show full SKILL.md (420 more words)Show less
2. README placement audit — by door rank (mandatory for user-facing features)

Adoption is a first-class goal and the README is the front door. For every change that ships something a user would care about, do not ask "where do I add a row?" — ask what is this capability's door rank (capability doors), and place it accordingly:

  • Rank 1–3, and it asks something of the reader (a practice loop) → the README's first screen: one sentence, its door, a link to its page. The first screen is bounded — something else has to leave.
  • Rank 1–3, ordinary capability → a headline entry in the README, a few lines, linked.
  • Rank 4–5 or configuration → one line in the generated capability reference (docs/reference/capabilities.md; regenerate with UPDATE_DOCS=1 npx jest capabilityIndex) plus its own docs page. The README does not grow.
  • New action → the actions page and the action count; no README row.
  • No user-facing surface (pure refactor, internal fix) → say so explicitly.

Never "add a row to the Features table" — that table was removed in #588 precisely because one row per epic produced a 69-row changelog in which nothing could be ranked. npm test enforces the ceiling (test/docs/readmeCeiling.test.ts), that nothing shipped disappeared (test/docs/readmeNamesKept.test.ts) and that this rule still reads this way (test/docs/placementRule.test.ts).

Blocking exit criterion, like the docs audit.

3. Walk the verification script (mandatory)

Open the issue's ## How to verify and do it — run every command in the automated table, then walk the manual steps in a real vault (npm run dev:vault). Two outcomes are failures, not paperwork:

  • A step does not describe reality → fix the spec section in the same change. A verification script nobody has walked is worth less than none, because it will be trusted.
  • A manual step turned out to be automatable → automate it and move it up into the table.

This is a blocking exit criterion (constitution §XIV): the change is not done until someone has seen it work by following the written script.

4. Quality check

Run the obsidian-plugin-quality skill and the obsidian-plugin-reviewer agent on the diff, and verify every acceptance criterion in the issue spec (body).

5. Close the issue via PR

Closing the issue (constitution §X). Commits only reference the issue ((#N)) — they never close it. The issue is closed by the pull request that merges the branch to main: put Closes #N (one line per addressed issue) in the PR body. Do not gh issue close from the feature branch and do not put closing keywords in commit messages. "Issue closed" is realised when the PR merges.

© RafaelGB, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in .claude/skills/implement of RafaelGB/Obsidian-ZettelFlow.

Open the folder on GitHubat commit 0676350

Compare with similar skills

Implement next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Implement compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Implement this skillRafaelGB/Obsidian-ZettelFlow174—~1.7kAutomated safety check: PassMIT
Obsidian CI Integrationjeremylongshore/tons-of-skills-marketplace2.8k—~2kAutomated safety check: PassMIT
Vulnhunter Fixcapitalone/VulnHunter1.1k1 repos~7.9kAutomated safety check: WarnApache-2.0
Auto GitHub Contributornexu-io/auto-github-contributor137—~2.5kAutomated safety check: NotesMIT
Gh IssuesTypedDevs/bashunit4341 repos~1.7kAutomated safety check: NotesMIT
Release Lambda LayerDataDog/datadog-lambda-js126—~1.8kAutomated safety check: PassApache-2.0

Similar skills

  • Obsidian CI Integration

    jeremylongshore/tons-of-skills-marketplace

    Set up GitHub Actions CI/CD for Obsidian plugin development.

    2.8k GitHub stars~2k tokensUpdated today
    DevOps & CloudAuto-check passed
  • Vulnhunter Fix

    capitalone/VulnHunter

    Automate vulnerability remediation from VulnHunter scan results using TDD.

    1.1k GitHub starsUsed in 1 repo~7.9k tokens
    Testing & QAAuto-check: warnings
  • Auto GitHub Contributor

    nexu-io/auto-github-contributor

    Interactively pick a quick-win contribution in any GitHub repo — either a labeled issue ("good first issue", "help wanted", docs) or a repo-scan candidate (typo, missing tests, i18n gap, actionable…

    137 GitHub stars~2.5k tokensUpdated 5 mo ago
    Testing & QAAuto-check: notes
  • Gh Issues

    TypedDevs/bashunit

    Walk over all open GitHub issues that are unassigned or assigned to the current user, and process each one via the /gh-issue skill, sequentially.

    434 GitHub starsUsed in 1 repo~1.7k tokens
    Testing & QAAuto-check: notes
  • Release Lambda Layer

    DataDog/datadog-lambda-js

    Official

    Walks through releasing a new datadog-lambda-js Lambda layer version — the automated Commercial release (version bump, tag, GitLab sign/publish jobs, npm publish, GitHub release) and the manual…

    126 GitHub stars~1.8k tokensUpdated yesterday
    Testing & QAAuto-check passed
  • Dev

    serithemage/serverless-openclaw

    Development workflow guide. An agent skill from serithemage/serverless-openclaw.

    196 GitHub stars~714 tokensUpdated 6 mo ago
    Testing & QAAuto-check: notes

More from RafaelGB/Obsidian-ZettelFlow

All 9 skills in this repo
  • New Action

    RafaelGB/Obsidian-ZettelFlow

    Scaffold a new ZettelFlow action following the project's 4-file convention (Action / Component / Settings / SettingsReader), register it, and document it.

    174 GitHub stars~1.1k tokensUpdated today
    Auto-check passed
  • Obsidian Plugin Quality

    RafaelGB/Obsidian-ZettelFlow

    Audit ZettelFlow against the official Obsidian plugin guidelines and the Community-hub automated review / quality score.

    174 GitHub stars~1.6k tokensUpdated today
    Auto-check passed
  • Release

    RafaelGB/Obsidian-ZettelFlow

    Cut a compliant ZettelFlow plugin release — version bump, versions.json, build, tag, and a GitHub Release whose notes follow the house format.

    174 GitHub stars~1.4k tokensUpdated today
    Auto-check passed
  • Specify

    RafaelGB/Obsidian-ZettelFlow

    Stage 1 of the SDD pipeline — write a ZettelFlow spec into a GitHub issue body (problem, value, functional requirements, testable acceptance criteria, capability disclosure, out-of-scope).

    174 GitHub stars~759 tokensUpdated today
    Auto-check passed
  • Sdd

    RafaelGB/Obsidian-ZettelFlow

    The Spec-Driven Development pipeline for ZettelFlow — the end-to-end flow (constitution → specify → plan → tasks → implement → verify) and which skill/agent owns each stage.

    174 GitHub stars~842 tokensUpdated today
    Auto-check passed
  • TDD

    RafaelGB/Obsidian-ZettelFlow

    The test-driven development workflow for ZettelFlow — write a failing test first, run jest, use the Obsidian mock and the bare-specifier alias mappings, and know which code is unit-testable.

    174 GitHub stars~724 tokensUpdated today
    Auto-check passed

Categories

Questions about Implement

What does Implement do?

Stage 4 of the SDD pipeline — execute a ZettelFlow task checklist (in a GitHub issue comment) test-first (red → green → refactor), one commit per advance, keeping npm run verify green and CI green…. Implement is an agent skill from RafaelGB/Obsidian-ZettelFlow. Stage 4 of the SDD pipeline — execute a ZettelFlow task checklist (in a GitHub issue comment) test-first (red → green → refactor), one commit per advance, keeping npm run verify green and CI green on every commit on a single feature branch.

When should I use Implement?

Implement fits situations like: tasks that involve Test-driven development.

How do I install Implement in Claude Code?

Run `npx skills add RafaelGB/Obsidian-ZettelFlow --skill implement -a claude-code`. Or copy the skill folder (.claude/skills/implement in RafaelGB/Obsidian-ZettelFlow) into .claude/skills/implement in your project. Claude Code loads it when a task matches its description.

How do I install Implement in Codex?

Run `npx skills add RafaelGB/Obsidian-ZettelFlow --skill implement -a codex`. Or copy the skill folder (.claude/skills/implement in RafaelGB/Obsidian-ZettelFlow) into .agents/skills/implement in your project. Codex loads it when a task matches its description.

Can I use Implement in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add RafaelGB/Obsidian-ZettelFlow --skill implement -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/implement, .gemini/skills/implement, .github/skills/implement and .opencode/skills/implement in your project.

What does Implement need to run?

Going by SKILL.md and its folder, Implement needs the command-line tools its instructions call (npm, gh, git and npx). Our summary lists: Node.js.

Does Implement access the network?

SKILL.md contains no URLs. Its commands use npm, gh, git and npx, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is Implement safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Implement use?

Implement is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Implement use?

About 1.7k tokens (SKILL.md is roughly 6.9k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Implement?

Skills that share tags, products or a category with Implement: Obsidian CI Integration (jeremylongshore/tons-of-skills-marketplace, 2.8k stars), Vulnhunter Fix (capitalone/VulnHunter, 1.1k stars), Auto GitHub Contributor (nexu-io/auto-github-contributor, 137 stars) and Gh Issues (TypedDevs/bashunit, 434 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Implement?

RafaelGB (a GitHub user) maintains it in RafaelGB/Obsidian-ZettelFlow, which has 174 GitHub stars. The repository holds 9 skills in this directory. The repository was last updated on October 8, 2026.

Source: RafaelGB/Obsidian-ZettelFlow on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.