CtrlK
BlogDocsLog inGet started
Tessl Logo

agent-sandbox

Agent skill for sandbox - invoke with $agent-sandbox

67

4.65x
Quality

52%

Does it follow best practices?

Impact

93%

4.65x

Average score across 3 eval scenarios

SecuritybySnyk

High

Do not use without reviewing

Fix and improve this skill with Tessl

tessl review fix ./.agents/skills/agent-sandbox/SKILL.md

The canonical home for this skill is agent-sandbox in ruvnet/claude-flow

SKILL.md
Quality
Evals
Security

Quality

Content

76%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 actionable and well-structured, giving Claude executable MCP calls and a clear deployment sequence. Its main weakness is the absence of validation and error-recovery feedback loops around destructive sandbox operations.

Suggestions

Add explicit validation checkpoints after sandbox creation and code execution, e.g. check sandbox_status before running code and verify execution output for errors before proceeding.

Include a feedback loop for failures (e.g. if execution errors, inspect logs, fix, re-run) around the delete/stop cleanup steps.

Tighten the "Quality standards" and "Your core responsibilities" prose into shorter imperative bullets to remove minor padding.

DimensionReasoningScore

Conciseness

The body is efficient with direct role framing, executable code, and a numbered workflow; it avoids explaining concepts Claude already knows, with only minor padding in the prose sections.

4 / 5

Actionability

It provides concrete, copy-paste-ready MCP calls with parameters, a language spec, and a complete template list covering the common cases, matching the fully-executable anchor.

5 / 5

Workflow Clarity

A clear six-step deployment sequence is present, but destructive/batch operations (create, execute, stop, delete sandbox) lack validation checkpoints and error-recovery feedback loops, which caps this dimension at 3 per the rubric guidance.

3 / 5

Progressive Disclosure

Content is well organized into clearly labeled sections (toolkit, deployment approach, templates, quality standards) with no nested references; minor organization gaps keep it just below the ideal single-file structure.

4 / 5

Total

16

/

20

Passed

Description

28%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 is too thin: it names the sandbox domain but provides no concrete capabilities, no natural trigger terms, and no explicit "Use when..." guidance. It relies on an invocation token for distinctiveness rather than descriptive content.

Suggestions

Add concrete actions, e.g. "Creates, configures, executes code in, and manages E2B sandbox environments."

Append a "Use when..." clause with natural user phrases such as "Use when the user asks to run code in an isolated sandbox, set up a development environment, or manage E2B sandboxes."

Include synonyms and concrete triggers (e.g. "E2B sandbox, isolated execution environment, dev sandbox") to improve trigger-term quality and distinctiveness.

DimensionReasoningScore

Specificity

The description "Agent skill for sandbox" names the domain but lists no concrete actions, matching the anchor where the domain is named but actions are minimal or generic.

2 / 5

Completeness

It gives a vague "what" (sandbox) with no "when/Use when..." trigger guidance, matching the anchor for a vague what and absent when.

2 / 5

Trigger Term Quality

Only the single keyword "sandbox" and the invocation token "$agent-sandbox" appear; there are no natural phrases a user would actually say, matching the one-generic-keyword anchor.

2 / 5

Distinctiveness Conflict Risk

"sandbox" alone is broad, but the explicit "$agent-sandbox" invocation token makes it somewhat specific, though it could still overlap with related sandbox skills.

3 / 5

Total

9

/

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.

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
ruvnet/ruflo
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.