Agent skill

Workfly

by automagik-dev in automagik-dev/genie

Discover a procedure and build its saved workflow — dynamic discovery, drafted script, adversarial verification, landed in the catalog.

MITAuto-check passedAgent Workflows

Install Workfly

skills CLI
$ npx skills add automagik-dev/genie --skill workfly -a claude-code

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

GitHub CLI
$ gh skill install automagik-dev/genie workfly --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/automagik-dev/genie.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/workfly .claude/skills/workfly && 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
workfly
GitHub stars
345
Token cost
~1.9k tokens
SKILL.md length
1,157 words
Files
2
Skills in repo
19
Repo updated
First seen
Licence
MIT

At a glance

Discover a procedure and build its saved workflow — dynamic discovery, drafted script, adversarial verification, landed in the catalog.

  • Works in 4 steps: Append readmeRow verbatim to the Entries… → Run the catalog's own contract suite… → When the objective converts an existing… → …
  • Agent Workflows work in your project
  • SKILL.md covers Invoke, Relay, Land it and Without a workflow surface, plus 1 more section
  • Calls bun

What it does

Workfly is an agent skill from automagik-dev/genie. Discover a procedure and build its saved workflow — dynamic discovery, drafted script, adversarial verification, landed in the catalog.

Its SKILL.md is about 1.9k tokens, which your agent loads only when the skill is triggered. The skill folder holds 2 other files (for example `agents/openai.yaml`).

It sits in Agent Workflows. The repository describes itself as: Wishes in, PRs out. CLI agent that interviews you, plans the work, dispatches parallel agents in isolated worktrees, and reviews code before you see it. The licence is MIT.

When your agent uses it

  • Agent Workflows work in your project

Example prompts

  • “/workfly”

Workflow steps

4 steps, taken from the first numbered list in SKILL.md.

  1. Append readmeRow verbatim to the Entries table of .claude/workflows/README.md, and check the result: that table names the new script on…
  2. Run the catalog's own contract suite yourself and read its output: it exits zero and its summary reports 0 fail before the work is called…
  3. When the objective converts an existing skill, rewrite that skill into a thin front door the way the council skill fronts its workflow: it…
  4. Report what the SPEC left open and every advisory finding nobody fixed. Silence there reads as "nothing was found", which is a different…

What it can do on your machine

Read from SKILL.md and the folder at commit c4f8788. 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:

    • bun

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

  • Network

    No URLs in SKILL.md.

    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

Workfly loads about 1.9k tokens when it runs. Until then it costs about 36 tokens; SKILL.md has 1,157 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~36
When it runs · the whole SKILL.md, loaded when a task matches
~1.9k

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 automagik-dev/genie at commit c4f8788, republished under its MIT licence (© automagik-dev). 1,157 words, ~1,927 tokens.

Download SKILL.mdSave it as .claude/skills/workfly/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.
name
workfly
description
Discover a procedure and build its saved workflow — dynamic discovery, drafted script, adversarial verification, landed in the catalog.
category
authoring
mutates
repo

Workfly

Use workfly when a procedure worth repeating — a skill, a runbook, a review ritual, a migration sweep — should become a saved workflow that any body runs unmodified. A procedure that needs the user in the middle of the run stays a skill; workfly encodes the part that can run on its own.

Workfly is a saved workflow, not a procedure this skill performs inline. The single source of truth is .claude/workflows/workfly.js in the genie repository's canonical workflow catalog (see .claude/workflows/README.md there); this skill is its front door. On a runtime that runs saved workflows, run the script at <repository root>/.claude/workflows/workfly.js when that file exists, otherwise ~/.claude/workflows/workfly.js (delivered by genie install / genie update); give the runtime that explicit script path, never a bare name — both scopes now carry these names and the order a name resolves in is undocumented. The orchestrator — the session model — specifies the objective and relays the result; the script is written by the workflow's own author agent, never drafted inline in the session.

Invoke

Pass {objective, sources?, name?, catalogDir?, model?, maxRepairs?, timestamp?} through args. Sources are repository-relative paths to whatever already describes the procedure — a runbook, a workflow README, or the review skill's own SKILL.md as it sits in this repository. name is kebab and the workflow checks it against every name already taken; catalogDir defaults to .claude/workflows and is the only directory the contract suite covers, so any other value comes back as a recorded coverage gap rather than a pass; maxRepairs is the repair budget — 2 when unset or not an integer, otherwise clamped to 0-3, so a larger value buys nothing; timestamp is the provenance stamp for the drafted script's header comment, and the workflow has no clock, so the caller is its only source. Everything else keeps its default.

text
args: {
  objective: "turn the nightly release-ledger sweep into a saved workflow",
  sources:   ["docs/_internal/release-ledger.md"],
  name:      "ledger-sweep",
  model:     "opus",
  timestamp: "2026-09-15T10:00:00Z"
}

Relay

A run that reached verification returns {ok, name, path, spec, readmeRow, frontDoor, parityGuard, verification, repairs, notConvened}. Relay verification unchanged: the static result, both refuters with their blocking and advisory findings intact, and ran, which names the lenses that answered — ok is never true while one of the three is silent. List notConvened — it holds exactly one kind of entry, an agent that returned nothing, counted as unverified and never as a pass; every other stop is reported through error.

A finding naming any file other than the drafted script — the catalog README row, the fronting skill, the parity test — always arrives as advisory, because those are the landing steps below and the repairer may edit only the script, so ok: false always names a defect in the script itself.

Read ok: false twice, because two shapes carry it. A run that stopped early returns {ok: false, error} and no verification: the objective was missing, a required reading did not answer or came back empty (no existing name reported is read as a failed reading, never as "every name is free"), the designer returned nothing or a SPEC with no name or a name that is not a single kebab slug, the author returned nothing, or the name was already taken. When the author returned nothing the error names the target path — check it for a partial write before re-running; every other early stop drafted nothing. Restate the objective, free the name, or re-run. A verified ok: false carries every surviving blocking finding in verification.blocking — including one the workflow synthesizes when a refuter answered refuted: true without naming a defect, so a bare refutation is never silent — and the static gate's own problems, which is where a non-default catalogDir reports that the contract suite never opened the drafted file. Read them, then decide whether to raise maxRepairs, change the objective, or land the fixes by hand.

Show full SKILL.md (544 more words)Show less

Land it

Once ok is true and the returned path holds a script:

  1. Append readmeRow verbatim to the Entries table of .claude/workflows/README.md, and check the result: that table names the new script on its own row.
  2. Run the catalog's own contract suite yourself and read its output: it exits zero and its summary reports 0 fail before the work is called done — in the genie repository that suite is bun test scripts/workflows-meta.test.ts, and a runner that names failing cases only leaves no per-case pass line to look for.
  3. When the objective converts an existing skill, rewrite that skill into a thin front door the way the council skill fronts its workflow: it states the script-path rule — the project file when the repository carries it, the ~/.claude/workflows/ copy otherwise, passed as the explicit script path either way — says what it relays unchanged and what stays with the user, keeps a by-hand fallback for a runtime with no workflow surface, and carries no roster of its own. Add a static parity test that pins that single-source roster, the way the council parity test pins council's.
  4. Report what the SPEC left open and every advisory finding nobody fixed. Silence there reads as "nothing was found", which is a different claim.

Without a workflow surface

Run the same five stages — Discover, Design, Draft, Verify, Repair — as subagents through the runtime's native delegation surface, with the briefs the script carries: three independent readers in parallel (source stages, catalog contract, token economy), one designer that merges the three readings into the SPEC, one author that writes the script file from the SPEC, then the static test and two refuters — semantics against the contract, fidelity against the source stages — both defaulting to refuted when uncertain and naming a blocking finding whenever they refute. Keep Repair bounded, re-verify after each round, and report what did not answer.

Rules

  • A saved name has to be free in both namespaces a runtime resolves from one list. The workflow checks the project catalog (.claude/workflows) and the skill namespace (skills/*) and returns any collision rather than overwriting anything — but a user-scope ~/.claude/workflows copy of the same name is outside what it can see, so confirm that scope by hand before landing (a stale user copy shadowed the genie repository's project council.js on 2026-09-15).
  • The workflow decides fan-out and per-agent effort; the front door states the objective and the sources, and may pin one model for the whole run. model stays a caller key on purpose — one model for the whole run is a front-door decision, per-agent tiers are not — and a front door that starts naming tiers has become a second source of truth.
  • Every conversion is one roadmap card, so the rewrite of the source skill and its parity test land together and stay reviewable.
  • Token economy is a criterion, not a slogan: a stage whose raw output would flood the orchestrator returns a summary rather than the raw result. That pricing is the economy reader's, and it shapes the SPEC; it is not carried back in the returned object.
  • A stage that needs the user mid-run stays in the front door. The refuters look for exactly that stage trapped inside a script, where the user can no longer reach it.

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

Files

SKILL.md and 1 other file in skills/workfly of automagik-dev/genie.

  • SKILL.md
  • agents/openai.yaml

Open the folder on GitHubat commit c4f8788

Compare with similar skills

Workfly 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.

Workfly compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Workfly this skillautomagik-dev/genie345—~1.9kAutomated safety check: PassMIT
MCP Server Builderanthropics/skills180k62 repos~2.3kAutomated safety check: PassApache-2.0
Hook Development for Claude Code Pluginsanthropics/claude-plugins-official37k11 repos~4.1kAutomated safety check: NotesApache-2.0
Using Superpowersfarm-fe/farm5.6k34 repos~1.4kAutomated safety check: PassMIT
Executing Plans Inlineobra/superpowers296k2 repos~5.1kAutomated safety check: PassMIT
Claude Code Agent Developmentanthropics/claude-plugins-official37k8 repos~2.8kAutomated safety check: PassApache-2.0

Similar skills

  • MCP Server Builder

    anthropics/skills

    Official

    Guides the design and implementation of Model Context Protocol servers in TypeScript or Python, from tool naming and error messages to evaluation.

    180k GitHub starsUsed in 62 repos~2.3k tokens
    Agent WorkflowsAuto-check passed
  • Hook Development for Claude Code Plugins

    anthropics/claude-plugins-official

    Official

    Explains how to write Claude Code plugin hooks, both prompt-based checks and bash commands, for events such as PreToolUse, Stop and SessionStart.

    37k GitHub starsUsed in 11 repos~4.1k tokens
    Agent WorkflowsAuto-check: notes
  • Using Superpowers

    farm-fe/farm

    A skill your agent uses when starting any conversation - establishes how to find and use skills, requiring Skill tool invocation before ANY response including clarifying questions

    5.6k GitHub starsUsed in 34 repos~1.4k tokens
    Agent WorkflowsAuto-check passed
  • Executing Plans Inline

    obra/superpowers

    Has the agent carry out an implementation plan itself, task by task in the current session, keeping a ledger, proving each step with a test and ending with one whole-branch review.

    296k GitHub starsUsed in 2 repos~5.1k tokens
    Agent WorkflowsAuto-check passed
  • Claude Code Agent Development

    anthropics/claude-plugins-official

    Official

    Explains how to write agents for Claude Code plugins: the markdown file with YAML frontmatter, trigger descriptions, model and color settings, and system prompt design.

    37k GitHub starsUsed in 8 repos~2.8k tokens
    Agent WorkflowsAuto-check passed
  • Skill Creator

    Azure/azqr

    Official

    Create new skills, modify and improve existing skills, and measure skill performance.

    794 GitHub starsUsed in 89 repos~8.2k tokens
    Agent WorkflowsAuto-check passed

More from automagik-dev/genie

All 19 skills in this repo
  • Learn

    automagik-dev/genie

    Diagnose and fix agent behavioral surfaces when the user corrects a mistake — connects to Claude native memory.

    345 GitHub stars~720 tokensUpdated yesterday
    Auto-check passed
  • Brainstorm

    automagik-dev/genie

    Explore an ambiguous idea with the user, settle scope and success criteria, and produce an independently reviewed design for wish.

    345 GitHub stars~2k tokensUpdated yesterday
    Auto-check passed
  • Report

    automagik-dev/genie

    Investigate a failure to its root cause with grounded evidence, hand the diagnosis to fix, and create a GitHub issue only when asked.

    345 GitHub stars~2k tokensUpdated yesterday
    Auto-check passed
  • Wish

    automagik-dev/genie

    Deliver one decided task end to end — admit it, work it in one worktree, gate, independent review, bounded repair, a merge-ready PR — or plan a multi-group wish when it is bigger than one task.

    345 GitHub stars~3.7k tokensUpdated yesterday
    Auto-check passed
  • Work

    automagik-dev/genie

    Execute an approved wish in dependency order with scoped workers, independent review, bounded repairs, and verified completion.

    345 GitHub stars~2.5k tokensUpdated yesterday
    Auto-check passed
  • Authoring

    automagik-dev/genie

    Write or revise a Genie skill so it survives the shipped contract — frontmatter, house size, starter card, and runtime-neutral voice.

    345 GitHub stars~1k tokensUpdated yesterday
    Auto-check passed

Categories

Questions about Workfly

What does Workfly do?

Discover a procedure and build its saved workflow — dynamic discovery, drafted script, adversarial verification, landed in the catalog. Workfly is an agent skill from automagik-dev/genie. Discover a procedure and build its saved workflow — dynamic discovery, drafted script, adversarial verification, landed in the catalog.

When should I use Workfly?

Workfly fits situations like: agent Workflows work in your project.

How do I install Workfly in Claude Code?

Run `npx skills add automagik-dev/genie --skill workfly -a claude-code`. Or copy the skill folder (skills/workfly in automagik-dev/genie) into .claude/skills/workfly in your project. Claude Code loads it when a task matches its description.

How do I install Workfly in Codex?

Run `npx skills add automagik-dev/genie --skill workfly -a codex`. Or copy the skill folder (skills/workfly in automagik-dev/genie) into .agents/skills/workfly in your project. Codex loads it when a task matches its description.

Can I use Workfly 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 automagik-dev/genie --skill workfly -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/workfly, .gemini/skills/workfly, .github/skills/workfly and .opencode/skills/workfly in your project.

What does Workfly need to run?

Going by SKILL.md and its folder, Workfly needs the command-line tools its instructions call (bun).

Does Workfly access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Workfly 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 Workfly use?

Workfly 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 Workfly use?

About 1.9k tokens (SKILL.md is roughly 7.7k 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 Workfly?

Skills that share tags, products or a category with Workfly: MCP Server Builder (anthropics/skills, 180k stars), Hook Development for Claude Code Plugins (anthropics/claude-plugins-official, 37k stars), Using Superpowers (farm-fe/farm, 5.6k stars) and Executing Plans Inline (obra/superpowers, 296k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Workfly?

automagik-dev (a GitHub organization) maintains it in automagik-dev/genie, which has 345 GitHub stars. The repository holds 19 skills in this directory. The repository was last updated on October 7, 2026.

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