CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

workflow

Creates durable, resumable workflows using Vercel's Workflow DevKit. Use when building workflows that need to survive restarts, pause for external events, retry on failure, or coordinate multi-step operations over time. Triggers on mentions of "workflow", "durable functions", "resumable", "workflow devkit", or step-based orchestration.

63

Quality

79%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./.agents/skills/workflow/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

68%Weight 40%Scale 1-5

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

A well-structured, mostly executable reference for the Workflow DevKit: concrete code, a compact limitations table, and clear pointers to bundled docs. The main gaps are the missing validation checkpoints in the build workflow, slight redundancy across the two sandbox sections, and a couple of code snippets that aren't fully self-contained.

Suggestions

Add an explicit validation checkpoint after framework setup (e.g., 'Run `npx workflow health` and only proceed once the endpoint responds') and reference it from the build guidance.

Merge "Prefer Step Functions to Avoid Sandbox Errors" and "Workflow Sandbox Limitations" into one section to remove the duplicated 'move to a step' advice.

Fix the streaming example (move `getWritable().close()` inside the step or show it as a complete function) and show the AI-SDK imports so snippets are copy-paste ready.

DimensionReasoningScore

Conciseness

The body is lean and assumes competence — code carries the weight ("'use step' functions have full Node.js access", a compact limitations table, tight error-handling and CLI snippets). Not a 5 because "Prefer Step Functions to Avoid Sandbox Errors" and "Workflow Sandbox Limitations" cover overlapping ground, and filler like "Benefits: Steps have automatic retry..." could be trimmed.

4 / 5

Actionability

Mostly executable guidance: complete step/workflow TypeScript examples, "globalThis.fetch = fetch", concrete FatalError/RetryableError branches, and copy-paste CLI commands ("npx workflow health", "npx workflow inspect runs"). Not a 5 due to minor gaps — the streaming snippet has a stray "await getWritable().close()" outside the step function, and processWithAI uses generateText/openai without showing their imports.

4 / 5

Workflow Clarity

The doc-lookup sequence ("Find docs: glob...", "Search content: grep...") is ordered, and debugging commands provide some checkpoints, but the overall build path has no validation step (e.g., 'run npx workflow health after wiring the endpoint' before proceeding). This matches the anchor 'steps listed but validation gaps; sequence present but checkpoints missing'.

3 / 5

Progressive Disclosure

Good structure: SKILL.md stays a distilled overview and defers bulk detail to clearly-signaled, one-level-deep bundled docs ("node_modules/workflow/docs/" with an explicit directory map, plus per-package docs paths). Not a 5 because reference material like the full supported-types serialization list and the imports quick reference is inlined where a references/ file would fit better at this length (~180 lines).

4 / 5

Total

15

/

20

Passed

Description

80%Weight 40%Scale 1-5

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

A strong description with explicit what/when structure, concrete capabilities, and explicit trigger terms. Its main weakness is that the bare trigger term "workflow" is highly generic and would fire on many non-DevKit workflow requests, creating real conflict risk with adjacent tooling.

Suggestions

Qualify the generic trigger (e.g., 'Triggers on mentions of Vercel Workflow DevKit, "durable functions", or "durable execution" when a Vercel/Next.js project is in use') to reduce overlap with CI/CD and generic agent workflows.

Add natural synonyms such as "durable execution", "long-running", or "pause and resume" to the trigger list to broaden natural-term coverage.

Name one or two additional concrete actions (e.g., 'start, inspect, and resume workflow runs via the client API') to move specificity from several actions to comprehensive.

DimensionReasoningScore

Specificity

"Creates durable, resumable workflows using Vercel's Workflow DevKit" plus "survive restarts, pause for external events, retry on failure, or coordinate multi-step operations over time" lists several concrete capabilities. Not a 5 because the actions are all facets of one operation (creating workflows) rather than a comprehensive list of distinct actions.

4 / 5

Completeness

Clearly answers both what ("Creates durable, resumable workflows using Vercel's Workflow DevKit") and when ("Use when building workflows that need to survive restarts, pause for external events, retry on failure..." plus explicit trigger mentions), in third-person voice. Both anchors of the 5 level are met explicitly.

5 / 5

Trigger Term Quality

Explicit triggers "workflow", "durable functions", "resumable", "workflow devkit", or "step-based orchestration" give good natural-term coverage for this domain. Not a 5 because common synonyms like "durable execution", "long-running", or "pause/resume" are missing.

4 / 5

Distinctiveness Conflict Risk

"Vercel's Workflow DevKit" and "durable functions" anchor a clear niche, but triggering on the bare word "workflow" overlaps heavily with CI/CD pipelines, GitHub Actions, and generic agent workflow requests. Somewhat specific, but the generic trigger keeps it at 3 rather than 4.

3 / 5

Total

16

/

20

Passed

Validation

100%

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

Kiz8-Team/open-agents

Path

.agents/skills/workflow/SKILL.md

Reviewed

Table of Contents

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.