CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

cloudbase-agent

Build and deploy AI agents with CloudBase Agent SDK (TypeScript & Python). Implements the AG-UI protocol for streaming agent-UI communication. Use when deploying agent servers, using LangGraph/LangChain/CrewAI adapters, building custom adapters, understanding AG-UI protocol events, or building web/mini-program UI clients. Supports both TypeScript (@cloudbase/agent-server) and Python (cloudbase-agent-server via FastAPI).

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 ./config/.claude/skills/cloudbase-agent/SKILL.md

The canonical home for this skill is cloudbase-agent in TencentCloudBase/CloudBase-AI-Toolkit

SKILL.md
Quality
Evals
Security

Quality

Content

56%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 clean, well-structured router whose routing logic (signal table, default rule, explicit destinations) is unambiguous, but in this bundle it dead-ends: the referenced language skill files do not exist, the missing-file fallback covers only sibling skills, and the core read-the-file instruction is repeated three times. The description is strong; the body's value depends entirely on bundle files that are not present.

Suggestions

Consolidate the 'read the language-specific file' directive — currently stated in the intro, in both Step 2 bullets, and again in the ⚠️ IMPORTANT block — into a single emphatic instruction to save tokens without losing enforcement.

Extend the missing-file fallback to cover ts/skill.md and py/skill.md (e.g., ask the user to install the full CloudBase plugin if a language file is absent), not just sibling skills, so the router never dead-ends.

Ship the ts/ and py/ skill files with the bundle, or inline a minimal quick start per language, so the router's referenced destinations actually exist in this environment.

DimensionReasoningScore

Conciseness

Mostly efficient but could be tightened: the core directive to read the language-specific file is stated three times (intro "then read the corresponding skill file", both Step 2 bullets "Read [ts/skill.md]… then follow ALL instructions", and the ⚠️ IMPORTANT block "you MUST read the corresponding skill file above"), and the sibling-skills preamble is meta-logistics rather than task guidance. Not anchor 4 because this repetition goes beyond minor trimming.

3 / 5

Actionability

The routing guidance itself is concrete (signal table, explicit default, exact file paths), but the action it directs — reading ts/skill.md or py/skill.md — dead-ends because neither file exists in the bundle, and the missing-file fallback covers only sibling skills, not these child files. Fits anchor 3 (some concrete guidance but incomplete, missing key details) rather than 4, whose guidance would be mostly executable as written.

3 / 5

Workflow Clarity

Clear two-step sequence (determine language via signal table with an enforced TypeScript default, then read the corresponding file) with an emphatic gate before code generation. Not anchor 5 because the failure mode where the ts/py destination file is missing is unhandled — a minor validation gap in an otherwise unambiguous workflow.

4 / 5

Progressive Disclosure

The design is a well-signaled one-level-deep router (SKILL.md → ts/skill.md / py/skill.md), which is the right structure, but scored against the actual bundle the referenced files are absent, so the navigation chain breaks immediately. Anchor 4's "minor organization gaps" understates entirely missing destinations, so 3 fits best.

3 / 5

Total

13

/

20

Passed

Description

96%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: concrete, third-person, explicitly separates what from when, and rich in natural trigger terms covering both language ecosystems. The only weakness is minor overlap risk from unqualified mentions of LangGraph/LangChain/CrewAI.

DimensionReasoningScore

Specificity

Lists multiple specific concrete actions — "Build and deploy AI agents", "Implements the AG-UI protocol for streaming agent-UI communication", "using LangGraph/LangChain/CrewAI adapters, building custom adapters", "building web/mini-program UI clients" — with concrete package names (@cloudbase/agent-server, cloudbase-agent-server via FastAPI). Coverage is comprehensive for the domain; it does not fit anchor 4 since no meaningful action is missing.

5 / 5

Completeness

Explicitly answers both what ("Build and deploy AI agents with CloudBase Agent SDK… Implements the AG-UI protocol for streaming agent-UI communication") and when ("Use when deploying agent servers, using LangGraph/LangChain/CrewAI adapters, building custom adapters, understanding AG-UI protocol events, or building web/mini-program UI clients") with concrete trigger phrases. Uses third-person voice throughout.

5 / 5

Trigger Term Quality

Natural trigger phrases a user would say are comprehensively covered: "deploying agent servers", "LangGraph/LangChain/CrewAI adapters", "AG-UI protocol events", "web/mini-program UI clients", plus synonyms across both ecosystems (TypeScript and Python package names). Not anchor 4 because no common variation or synonym is obviously missing.

5 / 5

Distinctiveness Conflict Risk

Clear niche (CloudBase Agent SDK + AG-UI protocol) with mostly distinct triggers, but "using LangGraph/LangChain/CrewAI adapters" could overlap with generic framework-specific skills. Fits anchor 4 (mostly distinct, minor overlap risk) better than 5 because the bare framework names are not qualified by CloudBase.

4 / 5

Total

19

/

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

frontmatter_unknown_keys

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

Warning

relative_links

Relative link issues: 2 missing

Warning

Total

14

/

16

Passed

Repository

TencentCloudBase/CloudBase-AI-Toolkit

Path

config/.claude/skills/cloudbase-agent/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.