write-flow

MUST use when creating flows.

Quality

50%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Medium

Suggest reviewing before use

Fix and improve this skill with Tessl

tessl review fix ./system_prompts/auto-generated/skills/write-flow/SKILL.md

Quality

Content

77%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is a strong, highly actionable Windmill flow-building guide with concrete commands, correct/incorrect examples, and clear sequenced workflows; its main weakness is the inlined schema dump and mild repetition that hurt token efficiency and file structure.

Dimension	Reasoning	Score
Conciseness	The core guidance is tight and high-signal (suffix resolution, whileloopflow state gotchas, branch scope rules), but the verbatim inline OpenFlow schema dump and repetition of generate-metadata/preview guidance across the CLI and "After writing" sections add tokens that do not all earn their place.	2 / 3
Actionability	It gives copy-paste-ready CLI commands (`wmill flow new … --summary`, `wmill flow preview --step`, `wmill generate-metadata --dry-run`) and complete correct/incorrect YAML plus an approval code sample, all fully executable.	3 / 3
Workflow Clarity	Flow creation is sequenced as explicit Steps 1–3 followed by a "Final Structural Self-Check" checklist, with validation checkpoints (offer preview before deploy, --dry-run before applying metadata) and clear intent-based preview/run/deploy decision rules.	3 / 3
Progressive Disclosure	The body is sectioned, but the full OpenFlow schema is inlined as a monolithic JSON wall that belongs in a separate reference, and references to OPENFLOW_SCHEMA and AGENTS.wmill.md are bare names rather than clearly signaled one-level-deep links to real files.	2 / 3
	Total	10 / 12 Passed

Description

22%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is overly terse: it captures a trigger but says nothing about what the skill does or that it targets Windmill, leaving it vague and at risk of conflicting with other flow-related skills.

Suggestions

Lead with concrete actions, e.g. "Scaffolds and writes Windmill flows via the wmill CLI — runs `wmill flow new`, fills in flow.yaml, and handles loops, branches, approvals, and preview/deploy."

Add natural trigger terms users would say: "windmill flow", "flow.yaml", "scaffold a flow", "wmill flow new".

Include an explicit "Use when…" clause tied to Windmill so it is distinguishable from generic flow/workflow skills.

Dimension	Reasoning	Score
Specificity	The description "MUST use when creating flows" states no concrete actions at all — it names neither scaffolding, CLI usage, flow.yaml editing, nor any capability, matching the vague/no-actions anchor rather than the domain-plus-actions level.	1 / 3
Completeness	It gives an explicit "when" ("use when creating flows") but entirely omits the "what does this do" half — per the guideline to score only what is explicitly stated, the missing what falls under the missing-what-or-when anchor.	1 / 3
Trigger Term Quality	"creating flows" is a phrase a user would naturally say, but coverage is minimal and omits common variations like "windmill flow", "flow.yaml", "scaffold a flow", or "wmill flow".	2 / 3
Distinctiveness Conflict Risk	"creating flows" is somewhat specific but, without a "windmill" qualifier, overlaps with other flow-related tools (Power Automate, flowcharts, generic workflows), so it could trigger for the wrong skill.	2 / 3
	Total	6 / 12 Passed

Validation

100%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: windmill-labs/windmill
Path: system_prompts/auto-generated/skills/write-flow/SKILL.md
Commit: 33d2845

Reviewed: 2 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.