CtrlK
BlogDocsLog inGet started
Tessl Logo

xai-agent-tools

xAI Agent Tools API for autonomous tool calling with X search, web search, and code execution. Use when building agents that need real-time data access and autonomous task execution.

55

Quality

69%

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 ./.claude/skills/xai-agent-tools/SKILL.md

The canonical home for this skill is xai-agent-tools in fernandezbaptiste/Skrillz

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.

The body is code-dense and mostly executable, covering configuration, patterns, costs, streaming, and errors in a logical order. Its main flaws are substantial redundancy across five near-identical agent examples, a missing demonstration of how tool configs are attached to a request, and no use of separate reference files for the pattern library.

Suggestions

Consolidate the five near-duplicate agent examples (Research, Analysis, Financial, Multi-Step, Cost-Optimized) into one or two representative patterns and move the rest to a references/agents.md file to cut significant token redundancy.

Show how the tool configurations are actually passed to the API (e.g., a tools parameter in the chat.completions.create call) so the Quick Start demonstrates enabling tools end-to-end.

Add validation guidance for tool outcomes — e.g., inspecting the response for tool results or errors before returning output — and integrate the error-handling pattern into the agent workflows rather than presenting it in isolation.

DimensionReasoningScore

Conciseness

Five near-identical sections ("Research Agent", "Analysis Agent", "Financial Agent", "Multi-Step Agent", plus "Cost-Optimized Agent") each repeat the same single API call with a different f-string prompt, and the star-rating model table adds little. This is noticeably verbose with several padded sections; it avoids anchor 1 only because there is no over-explanation of basic concepts and much of the code is genuinely useful.

2 / 5

Actionability

Nearly all code is copy-paste executable (client setup, chat.completions calls, streaming, error handling, conversational class) with concrete tool config dicts and a cost table. It falls short of anchor 5 because the tool configurations (x_search_config, web_search_config, code_execution_config) are never wired into an API call — no request parameter showing how tools are actually enabled is ever demonstrated.

4 / 5

Workflow Clarity

The body follows a logical progression (Quick Start → Tool Configurations → Agent Patterns → Cost Management → Error Handling) and includes checkpoints such as the try/except error-handling example that returns structured success/error results. It misses anchor 5 because error handling is shown in isolation and there is no guidance on validating or inspecting tool outcomes in responses.

4 / 5

Progressive Disclosure

Sections are clearly headed and navigable with external doc links listed, but all ~330 lines live in a single SKILL.md with no bundle files; the library of agent pattern examples clearly belongs in a separate references file. This matches anchor 3 (some structure, content that should be separate is inline) rather than anchor 4 (no internal split or one-level-deep reference signaling exists).

3 / 5

Total

13

/

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 solid description that clearly states both what the skill covers and when to use it, in third person, with good natural trigger terms. Its main weaknesses are a label-like 'what' clause that names tools rather than concrete actions, and missing common synonyms like Grok or Twitter/X posts.

DimensionReasoningScore

Specificity

The description names the domain and tool set ("autonomous tool calling with X search, web search, and code execution") but describes the API rather than enumerating concrete actions the skill performs. It sits above anchor 2 (more than a bare domain label) but below anchor 4 (no list of specific actions such as searching posts or executing Python).

3 / 5

Completeness

Both parts are explicit: what ("xAI Agent Tools API for autonomous tool calling with X search, web search, and code execution") and when ("Use when building agents that need real-time data access and autonomous task execution"). The 'what' is somewhat label-like and the 'when' is a single clause without broader trigger variants, keeping it below the fully concrete anchor 5.

4 / 5

Trigger Term Quality

Natural phrases users would say are present: "building agents", "real-time data access", "X search", "web search", "code execution", "autonomous task execution". Missing a few natural variations users might say ("Grok", "Twitter/X posts", "agentic"), so it falls short of the comprehensive synonym coverage of anchor 5 but is clearly above anchor 3.

4 / 5

Distinctiveness Conflict Risk

"xAI Agent Tools API" carves out a clear niche with distinct triggers around server-side agentic tool calling. Minor overlap risk remains with closely related xAI skills (e.g., a dedicated X-search skill would also trigger on "X search"), keeping it below the minimal-conflict anchor 5.

4 / 5

Total

15

/

20

Passed

Validation

93%

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

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

frontmatter_unknown_keys

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

Warning

Total

15

/

16

Passed

Repository
fernandezbaptiste/Skrillz
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.