wf-explain

A read-only explainer for workflow scripts, which are programs that automate a sequence of tasks. It describes their control flow, inputs, costs, requirements, safety implications, and possible effects on a repository without running them.

In plain words
What is it for?
Use it when someone gives you a workflow script or when you want to inspect one by name or path. It can also produce a static structure diagram of the script.
Why use it?
It lets you inspect an unfamiliar automation before granting it permission to run. This helps reveal what the script needs and what it may change.

Command

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 commands/hesreallyhim/ultracode-workflows/wf-explain
Clone the repo
git clone --depth 1 https://github.com/hesreallyhim/ultracode-workflows
Per session 27 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 960 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 $0.00027 $0.00960
Opus 5 $0.00014 $0.00480
Sonnet 5 $0.00005 $0.00192
Haiku 4.5 $0.00003 $0.00096

Measured 2d ago against content hash 690514717896, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

wf-explain 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.

plugins/ultracode-workflows/commands/wf-explain.md · 45 lines

How it starts

The opening of the file, as written. The whole thing — 45 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/wf-explain

Read the workflow script the user named in $ARGUMENTS and explain it in plain language: what it does, how it's wired, what it needs, what it costs, and what it will do to their repo. This is the command to run before executing an unfamiliar workflow — especially one someone else wrote.

Resolve the target: a path if given; a bare name means ${CLAUDE_PLUGIN_ROOT}/workflows/<name>.js, falling back to .claude/workflows/<name>.js in the current project.

Do not execute the script

Analyze without running the workflow harness. A workflow script is ordinary JavaScript that runs with the user's full privileges, and this command exists precisely for scripts whose trustworthiness is unknown.

You may use the bundled diagram tool as a source-analysis aid:

node "${CLAUDE_PLUGIN_ROOT}/scripts/diagram-workflow.mjs" <resolved-path>

This prints a terminal diagram. If you relay that diagram to the user, put it inside a fenced plain-text code block so its box-drawing columns remain aligned. diagram-workflow.mjs --structure <resolved-path> prints structured static evidence instead. Neither mode evaluates the workflow body. Use --md only when the user explicitly requests Mermaid for a document or another Mermaid-capable destination; Claude Code's TUI does not render it.

The diagram reports exact, ranged, or unknown multiplicities according to what the source proves; it does not consume runtime args or explore one sampled execution path. Its output is structural evidence, not a safety verdict, so still read the source with Read/Grep and complete every check below. Do not invoke the workflow itself until that review supports doing so.

Explain these six things

  1. Purpose — what it produces, in one or two sentences, and when someone would reach for it.

  2. Control flow — walk the phases in order (meta.phases and phase() calls). For each: what fans out and how wide, whether concurrency comes from parallel() (a barrier — waits for all) or pipeline() (per-item, no barrier), and any loops with their termination condition and caps. Name the orchestration patterns you recognize (adversarial verification, multi-modal sweep, judge panel, loop-until-dry, manifest sharding, checkpoint) so the user learns the vocabulary.

Read the full file on GitHub · 45 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 · 45 lines · 27 tokens per session scan A 690514717896

Subscribe to this mod's changes

wf-explain is a command published in the GitHub repository hesreallyhim/ultracode-workflows (2 stars, last pushed 26d ago), licensed MIT. It adds 27 tokens to every session and 960 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 2026-08-31.