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
Evaluation results
90%
68%Automated Data Analysis Pipeline
Python sandbox lifecycle management
Python template
0%
100%
Timeout set
0%
100%
Correct create function
0%
100%
File upload used
0%
100%
Correct execute function
0%
100%
capture_output enabled
0%
100%
Env vars for secrets
50%
100%
Sandbox cleanup
0%
100%
Status check
0%
0%
Error handling
100%
77%
Deployment order
100%
100%
Without context: $0.4100 · 1m 54s · 21 turns · 27 in / 6,400 out tokens
With context: $0.3891 · 1m 28s · 22 turns · 27 in / 4,913 out tokens
100%
81%Node.js Microservice Test Harness
Node.js sandbox with package installation and status monitoring
Node template
0%
100%
install_packages used
0%
100%
Correct create function
0%
100%
Correct execute function
0%
100%
capture_output enabled
0%
100%
Status monitoring
0%
100%
Timeout specified
0%
100%
Sandbox cleanup always runs
0%
100%
Error handling present
100%
100%
Execution logging
100%
100%
Language parameter
0%
100%
Without context: $0.3561 · 1m 31s · 17 turns · 24 in / 5,668 out tokens
With context: $0.4118 · 1m 37s · 21 turns · 269 in / 5,914 out tokens
91%
72%Cross-Platform Application Smoke Test
Multi-environment sandbox orchestration and cleanup
React template used
0%
100%
Python template used
0%
100%
Multiple sandbox_create calls
0%
100%
Correct create function
0%
100%
Status checked per environment
0%
0%
Execute in each environment
0%
100%
capture_output on executions
0%
100%
All sandboxes cleaned up
0%
100%
Cleanup despite failures
70%
100%
Timeout on each sandbox
0%
100%
Execution logging
100%
100%
Parallel or sequential orchestration
100%
100%
Without context: $0.5080 · 2m 26s · 20 turns · 25 in / 10,147 out tokens
With context: $0.4957 · 2m 7s · 22 turns · 377 in / 7,704 out tokens
- Evaluated
- Agent
- Claude Code
- Model
- Unknown
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.