wf-compose

A workflow-planning tool that reads the conversation, checks available skills, finds missing pieces, and proposes a reusable Weft workflow with repeated steps.

In plain words
What is it for?
Use it to create a workflow from a request, ask for details interactively, or adjust an existing workflow template.
Why use it?
It helps turn a loosely described task into an organised process, so important steps and available tools are less likely to be missed.

Skill for Claude CodeCodex

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/dioptx/weft/wf-compose
Any agent
npx skills add dioptx/weft --skill wf-compose
Clone the repo
git clone --depth 1 https://github.com/dioptx/weft

Made for: Claude Code, Codex.

Per session 25 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,495 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00025 $0.01495
Opus 5 $0.00013 $0.00747
Sonnet 5 $0.00005 $0.00299
Haiku 4.5 $0.00003 $0.00150

Measured 2d ago against content hash 4b0fe18e93c2, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

wf-compose scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/wf-compose/SKILL.md · 167 lines

How it starts

The opening of the file, as written. The whole thing — 167 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Compose a Weft Workflow

Read the conversation context, scan available skills, identify gaps, and propose a v2 workflow template with loops and skill blocks.

Arguments

$ARGUMENTS

Modes

Usage Behavior
/wf-compose "review, fix, iterate until clean" One-shot: propose from description
/wf-compose (no args) Interactive: ask "What are you trying to accomplish?"
/wf-compose --from feature-workflow Start from existing template, modify based on context

Step 1: Gather Context

Understand what the user is trying to do:

  1. Review the recent conversation for intent (what task, what repo, what outcome).
  2. Check git state:
    git branch --show-current 2>/dev/null
    git diff --stat 2>/dev/null | tail -5
    
  3. Check if a weft workflow is already active:
    python3 "${CLAUDE_PLUGIN_ROOT}/core/cli.py" status 2>/dev/null
    
  4. If --from <template> was provided, load it as the starting point:
    python3 "${CLAUDE_PLUGIN_ROOT}/core/cli.py" preview <template>
    

Step 2: Scan Skill Registry

Build a map of what skills are available:

  1. Read the local skills registry, if any (path varies by setup):
    cat "${CLAUDE_SKILLS_REGISTRY:-$HOME/.claude/skills-registry.json}" 2>/dev/null
    
  2. List weft templates:
    python3 "${CLAUDE_PLUGIN_ROOT}/core/cli.py" start
    
  3. Categorize skills by function (examples — substitute what you have available):
    • Review: staff-review, arch-review, code-review, differential-review
    • Fix/Polish: fix-polish, refactor, simplify
    • Test: infra-test, webapp-testing
    • Plan: aot-plan, spec-first
    • Research: perplexity, context7, research-loop
    • Deploy: deploy-service, pr-ready

Step 3: Gap Analysis

Compare what the user described against available skills:

  1. Extract skill references from the user's description (explicit names like "/staff-review" or implicit like "review code", "test it", "deploy").
  2. For each referenced skill, check if it exists in the registry.
  3. For missing skills, present options:
    Missing skill: /devils-advocate
    Options:
    1. Create a stub skill (I'll generate a skeleton)
    2. Use /staff-review instead (similar purpose)
    3. Skip this step
    
  4. Wait for user choice on each gap before proceeding.

Read the full file on GitHub · 167 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 167 lines · 25 tokens per session scan A 4b0fe18e93c2

Subscribe to this mod's changes

wf-compose is a skill published in the GitHub repository dioptx/weft (22 stars, last pushed 17d ago), licensed MIT. It adds 25 tokens to every session and 1,495 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.