agent-sandbox
Agent skill for sandbox - invoke with $agent-sandbox
55
Does it follow best practices?
If you maintain this skill, you can automatically optimize it using the tessl CLI to improve its score:
npx tessl skill review --optimize ./path/to/skillEvaluation — 93%
↑ 4.65xAgent success when using this skill
Validation for skill structure
Discovery
0%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This description is critically deficient across all dimensions. It provides no concrete actions, no natural trigger terms, no usage guidance, and no distinguishing characteristics. Claude would have no basis for selecting this skill appropriately from a collection of skills.
Suggestions
Add specific concrete actions describing what this skill does (e.g., 'Executes code in isolated sandbox environments, manages sandbox sessions, runs untrusted scripts safely')
Add a 'Use when...' clause with natural trigger terms users would say (e.g., 'Use when the user asks to run code safely, test scripts, execute in isolation, or mentions sandbox/sandboxed execution')
Remove the invocation syntax '$agent-sandbox' from the description as it's implementation detail, not selection criteria
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description contains no concrete actions whatsoever. 'Agent skill for sandbox' is completely abstract with no indication of what the skill actually does. | 1 / 3 |
Completeness | The description fails to answer 'what does this do' (beyond vague 'agent skill') and provides no 'when should Claude use it' guidance. Both components are missing or extremely weak. | 1 / 3 |
Trigger Term Quality | The only potential trigger term is '$agent-sandbox' which is technical jargon, not a natural keyword users would say. No natural language terms are provided. | 1 / 3 |
Distinctiveness Conflict Risk | 'Agent skill for sandbox' is extremely generic and provides no distinguishing characteristics. It could conflict with any sandbox-related or agent-related skill. | 1 / 3 |
Total | 4 / 12 Passed |
Implementation
64%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The skill provides good actionable code examples for E2B sandbox management with clear API usage patterns. However, it suffers from unnecessary persona framing, lacks validation checkpoints in the workflow, and could be more concise by removing obvious responsibilities that Claude would naturally understand.
Suggestions
Remove the persona introduction and 'core responsibilities' section - these describe what Claude already knows how to do
Add explicit validation steps to the workflow, such as checking sandbox_status after creation before executing code
Condense the 'Quality standards' section into the workflow itself as inline checkpoints rather than a separate list
Add error handling examples showing what to do when sandbox creation or execution fails
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The content includes some unnecessary framing ('You are a Flow Nexus Sandbox Agent') and lists responsibilities/standards that Claude already understands. The toolkit section is efficient, but the surrounding context adds padding. | 2 / 3 |
Actionability | Provides concrete, executable JavaScript code examples for all major operations (create, execute, upload, status, stop, delete). The API calls are copy-paste ready with clear parameter structures. | 3 / 3 |
Workflow Clarity | The 6-step deployment approach provides a sequence but lacks explicit validation checkpoints. No feedback loops for error recovery or verification steps between sandbox creation and code execution. | 2 / 3 |
Progressive Disclosure | Content is reasonably organized with clear sections, but everything is inline in a single file. Template descriptions and quality standards could be separated or condensed. No references to external documentation. | 2 / 3 |
Total | 9 / 12 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 — 11 / 11 Passed
Validation for skill structure
No warnings or errors.
- 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.