CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

cloudflare-workflows

Cloudflare Workflows for durable long-running execution. Use for multi-step workflows, retries, state persistence, or encountering NonRetryableError, execution failed errors.

57

Quality

72%

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 ./plugins/cloudflare-workflows/skills/cloudflare-workflows/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

63%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, code-first body with a strong quick start, explicit error/fix feedback loops, and clearly signaled one-level-deep references. Its two real defects are broken navigation — templates/, scripts/, commands, and agents are referenced (and counted in frontmatter metadata) but absent from the bundle — and redundant sections (Troubleshooting duplicating the Top-5 errors, tooling sections for files that don't exist).

Suggestions

Remove or merge the "Troubleshooting" section into "Top 5 Critical Errors" (it restates the same five errors and the same common-issues.md pointers), and fold the thin "Wrangler Commands" section into the "When to Load References" entry — the body currently covers those five errors twice.

Make the bundle match the body: either ship the 8 templates/ and 5 scripts/ files (plus the commands/agents) that are referenced throughout, or delete those sections and the "Load templates/... for complete example" pointers so no referenced path 404s.

Trim the "Commands", "Agents", and "Scripts" tables (and the matching frontmatter counts like templates_included/scripts_included) to only what actually exists in the skill bundle.

DimensionReasoningScore

Conciseness

Most sections (Quick Start, Critical Rules, Top 5 errors) are tight and code-dense, but there is real duplication and padding: the "Troubleshooting" section restates the exact same five errors already fully covered in "Top 5 Critical Errors" (both pointing to common-issues.md), "Wrangler Commands" duplicates the reference-list entry, "State Persistence" is a concept demo, and the Commands/Agents/Scripts sections document tooling that does not exist in the bundle. This fits 'mostly efficient but includes some unnecessary explanation or could be tightened' — not 4 because the duplicated troubleshooting content and phantom-tooling sections are more than minor trimmable instances, and not 2 because the majority of the body is efficient.

3 / 5

Actionability

Every core task has complete, executable inline code: scaffold command, full WorkflowEntrypoint class, wrangler.jsonc config, deploy command, step-method signatures, and ❌/✅ fix pairs for each documented error. It is not 5 because a substantial set of instructions cannot actually be run — the usage lines for `./scripts/*.sh` (5 scripts), the 8 "Load templates/..." pointers, and the slash commands reference files absent from the bundle; it stays above 3 because the inline code is self-contained and copy-paste ready for all common cases.

4 / 5

Workflow Clarity

The Quick Start is a clear numbered sequence (create → structure → configure → deploy), each of the five error sections pairs symptom → cause → wrong/right fix (a concrete feedback loop), and "When to Load References" maps symptoms to files. Minor validation gap: the deploy step has no inline verification (it is deferred to the production-checklist reference). Matches 'Clear sequence with most checkpoints present; minor validation gaps' — not 5 for the missing deploy-verification step, not 3 because sequences and error-recovery paths are explicit.

4 / 5

Progressive Disclosure

Good progressive-disclosure structure: an overview-style body with bulk detail deferred to 8 references/*.md files that all exist, one level deep, and an exemplary "When to Load References" section giving per-file trigger conditions. Not 5 (clear overview with well-signaled one-level-deep references) because navigation is broken for a large share of paths — all 8 templates/*, all 5 scripts/*, the 4 slash commands, and the 3 agents referenced in the body (and claimed in frontmatter metadata) do not exist in the bundle, so "Load templates/basic-workflow.ts" and `./scripts/validate-workflow-config.sh` are dead ends; not 3 because the signaling and placement of the content that does exist is well above 'could be better organized'.

4 / 5

Total

15

/

20

Passed

Description

70%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 compact, well-targeted description with an explicit "Use for..." trigger clause and product-specific error tokens that make it highly discoverable. Its main weakness is that it never states what the skill concretely does — the capability side is a single domain phrase rather than actions.

DimensionReasoningScore

Specificity

The description names the domain precisely ("Cloudflare Workflows for durable long-running execution") but lists no concrete actions the skill performs — no build/configure/deploy/debug verbs; "multi-step workflows, retries, state persistence" are trigger situations rather than capabilities. This matches the anchor 'Names domain and 1-2 concrete actions, but not comprehensive' — not 4 because there is no list of specific actions, and not 2 because the domain framing is concrete rather than generic.

3 / 5

Completeness

Both halves are present: the "what" is "Cloudflare Workflows for durable long-running execution" and the "when" is an explicit "Use for..." clause with concrete trigger phrases. Not 5 because the "what" stays at domain level without stating what the skill actually does (scaffold, configure, debug); not 3 because the "Use for..." clause is explicit rather than weakly implied.

4 / 5

Trigger Term Quality

"multi-step workflows, retries, state persistence, or encountering NonRetryableError, execution failed errors" includes exact error strings a developer would paste ("NonRetryableError", "execution failed"), giving good natural keyword coverage. It falls short of 5 only for missing common variations like "Cloudflare Workers", "step.do", or "stuck workflow"; it exceeds 3 because coverage goes beyond a couple of generic keywords.

4 / 5

Distinctiveness Conflict Risk

"Cloudflare Workflows" plus the product-specific "NonRetryableError" token forms a clear niche with distinct triggers and minimal conflict risk. Not 5 because the generic tail "multi-step workflows, retries" could also match non-Cloudflare orchestration requests (e.g., CI pipelines); not 3 because the product-specific tokens anchor it well against most similar skills.

4 / 5

Total

15

/

20

Passed

Validation

87%

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

Validation — 14 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

skill_md_line_count

SKILL.md is long (576 lines); consider splitting into references/ and linking

Warning

metadata_field

'metadata' should map string keys to string values

Warning

Total

14

/

16

Passed

Repository

secondsky/claude-skills

Path

plugins/cloudflare-workflows/skills/cloudflare-workflows/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.