CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

workflow

Vercel Workflow SDK expert guidance. Use when building durable workflows, long-running tasks, API routes or agents that need pause/resume, retries, step-based execution, or crash-safe orchestration with Vercel Workflow.

Quality

76%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Critical

Do not install without reviewing

Fix and improve this skill with Tessl

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

Quality

Content

71%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 dense, highly actionable reference for a novel SDK, with excellent executable code and useful validation guidance. Its weaknesses are repetition (duplicated error-handling sections, redundant streaming examples) and a monolithic structure that inlines API reference material a bundled reference file should hold.

Suggestions

Merge the 'Error handling' and 'Error handling patterns' sections into one, keeping the FatalError/RetryableError/Promise.allSettled strategy table and a single code example.

Collapse the four near-identical namespaced-stream writer functions into one generic pattern (e.g., a makeLogger(namespace) factory) and trim the consuming example to the essential namespace accesses.

Move the World SDK method signatures, event-type tables, and extended streaming examples into a bundled reference file (e.g., references/world-sdk.md) and keep a short pointer plus 2-3 key examples in SKILL.md.

DimensionReasoningScore

Conciseness

Mostly efficient, high-novelty SDK-specific guidance, but with clear tightening opportunities: FatalError/RetryableError behavior is explained in two separate sections ('Error handling' and 'Error handling patterns'), and the namespaced-streams section includes four near-identical writer functions (logDebug, logInfo, emitAgentThought, emitAgentResult) plus a long consuming example that one generalized pattern would cover. This matches 'mostly efficient but includes some unnecessary explanation or could be tightened' rather than the level-4 'minor instances'.

3 / 5

Actionability

Fully executable, copy-paste-ready TypeScript and CLI commands throughout: complete step-function, WorkflowAgent, start(), hook, serde-protocol, streaming, and Vitest examples, plus concrete npx workflow commands with flags. Specific examples cover the common cases, matching the level-5 anchor.

5 / 5

Workflow Clarity

The body is organized into clear topical sections with concrete directives, and validation checkpoints appear where they matter (serde compliance via 'workflow transform --check-serde' and 'workflow validate', 'npx workflow health', explicit unit-vs-integration test separation). As a reference-style skill there is no single end-to-end sequence, and a few checkpoints are implicit, so it fits 'clear sequence with most checkpoints present; minor validation gaps' rather than the level-5 explicit-feedback-loop anchor.

4 / 5

Progressive Disclosure

No bundle reference files exist (no references/, scripts/, or assets/ directories), while hundreds of lines of API-reference material — World SDK method signatures, event-type tables, and the extended namespaced-streaming examples — are inlined in SKILL.md. Pointers to node_modules/*/docs/ are clearly signaled, but content that should live in separate reference files is inline, matching 'some structure; content that should be separate is inline' at level 3.

3 / 5

Total

15

/

20

Passed

Description

75%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 solid description with a clear niche, an explicit use-when clause, and good natural trigger coverage. Its main weakness is the vague 'expert guidance' framing for the what, and a handful of missing common trigger synonyms.

Suggestions

Replace the abstract 'Vercel Workflow SDK expert guidance' opening with concrete actions the skill performs, e.g., 'Build durable workflows with step-based execution, pause/resume hooks, retries, and crash-safe orchestration using the Vercel Workflow SDK.'

Add a few high-frequency natural trigger phrases to the when-clause, such as 'step functions', 'webhooks', or 'durable execution', to broaden matching without changing the niche.

DimensionReasoningScore

Specificity

Names the domain ("Vercel Workflow SDK") and lists several specific capability areas — "durable workflows, long-running tasks, API routes or agents that need pause/resume, retries, step-based execution, or crash-safe orchestration" — but the core action "expert guidance" is abstract and doesn't state what the skill concretely does. This fits 'several specific actions; minor gaps' better than the comprehensive level 5 and is clearly above the 1-2-action level 3.

4 / 5

Completeness

Both what and when are present, with an explicit "Use when building..." clause containing concrete triggers. The 'what' ("Vercel Workflow SDK expert guidance") is generic compared to the level-5 anchor's concrete action list, matching 'both present; when could be more explicit or specific' inverted — the when is strong but the what could be more specific.

4 / 5

Trigger Term Quality

Good coverage of natural phrases users would say: "durable workflows", "long-running tasks", "pause/resume", "retries", "crash-safe orchestration". A few common variants are missing (e.g., "step functions", "webhooks", "durable execution"), so it sits between 'some relevant keywords' (3) and 'comprehensive including synonyms' (5).

4 / 5

Distinctiveness Conflict Risk

Both sentences are anchored to "Vercel Workflow", giving a clear niche with low conflict risk. Generic terms like "API routes", "agents", and "orchestration" carry minor overlap risk with CI-workflow or agent-orchestration skills, keeping it below the 'minimal conflict risk' level 5.

4 / 5

Total

16

/

20

Passed

Validation

75%

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

Validation — 12 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

skill_md_line_count

SKILL.md is long (1258 lines); consider splitting into references/ and linking

Warning

metadata_version

'metadata.version' is missing

Warning

metadata_field

'metadata' should map string keys to string values

Warning

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

12

/

16

Passed

Repository

vercel/vercel-plugin

Path

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.