CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

agents-sdk

Build, debug, or review Cloudflare Agents SDK applications using the agents package.

60

Quality

76%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./config/claude/skills/agents-sdk/SKILL.md

The canonical home for this skill is agents-sdk in cloudflare/skills

SKILL.md
Quality
Evals
Security

Quality

Content

86%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.

The body is a lean, highly actionable retrieval-first reference: executable config and code, a complete API syntax table, and a well-organized one-level-deep bundle with all referenced files present. Its only real weakness is redundancy between the retrieval table, the Capabilities list, and the References section.

DimensionReasoningScore

Conciseness

The body is dense and efficient (terse tables, command blocks, gotcha lists) with no re-teaching of known concepts, but the 18-bullet Capabilities list substantially duplicates the retrieval table's 'Use for' column and the closing References section re-lists the same topics, which could be trimmed.

4 / 5

Actionability

Copy-paste ready throughout: install commands with expected output ('npm ls agents # Should show agents package'), a complete wrangler.jsonc block, a fully runnable Counter class with imports and default export, and a Core APIs table with exact syntax for every common task, plus concrete gotchas like 'Do NOT enable experimentalDecorators'.

5 / 5

Workflow Clarity

'FIRST: Verify Installation' gives an explicit starting step with a fallback path, and the 'Prefer retrieval over pre-training' directive plus the topic table makes the retrieve-then-implement sequence navigable; however, the sequence is implied by section order rather than explicitly numbered, with no stated verify step after implementation.

4 / 5

Progressive Disclosure

Verified against the actual bundle: all 19 referenced references/*.md files exist, are one level deep (cross-links inside references are sibling pointers, not nesting), and are grouped under labeled categories (Core, Chat & Streaming, Background Processing, Integrations, Experimental) with one-line descriptions — a clear overview with well-signaled, easy-to-navigate references.

5 / 5

Total

18

/

20

Passed

Description

53%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.

The description clearly identifies the domain and distinguishes the skill from generic tooling, but it omits any 'when to use' guidance and relies on broad verbs rather than naming concrete capabilities or natural trigger synonyms. It sits just above the midpoint: distinct and honest, yet not self-activating.

Suggestions

Add an explicit 'Use when...' clause, e.g. 'Use when building, debugging, or reviewing Cloudflare Agents SDK apps, Durable Object agents, or Cloudflare chat agents.'

Name concrete capabilities to raise specificity, such as 'persist state with setState, schedule cron tasks, build streaming chat agents, or expose RPC via @callable'.

Include natural trigger terms users would say: 'Durable Objects', 'AIChatAgent', 'useAgent', 'agents SDK', 'Cloudflare agent'.

DimensionReasoningScore

Specificity

Names the domain ('Cloudflare Agents SDK applications using the agents package') and lists actions ('Build, debug, or review'), but these are generic verbs with no concrete capabilities named (state, scheduling, MCP, chat), matching the anchor for domain plus 1-2 actions rather than several specific actions.

3 / 5

Completeness

The 'what' is clear (build, debug, or review Agents SDK applications), but there is no 'Use when...' clause or equivalent trigger guidance, which caps completeness at 3 per the judging guidelines.

3 / 5

Trigger Term Quality

Includes natural phrases like 'Cloudflare Agents SDK', 'agents package', and 'build/debug/review', but misses common variations a user would say such as 'Durable Objects', 'AIChatAgent', 'useAgent', or 'Cloudflare agent', matching the anchor for relevant keywords missing common synonyms.

3 / 5

Distinctiveness Conflict Risk

'Cloudflare Agents SDK' plus 'agents package' pins a distinct niche with specific triggers, but agent-building is a crowded space (other agent SDKs/frameworks), leaving minor overlap risk with closely related skills.

4 / 5

Total

13

/

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

freekmurze/dotfiles

Path

config/claude/skills/agents-sdk/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.