CtrlK
BlogDocsLog inGet started
Tessl Logo

sandbox-exec-fallback

Fallback pattern for executing Python code when execute_code_sandbox fails by writing to file and running via shell

48

Quality

50%

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 ./benchmarks/gdpval/skills/sandbox-exec-fallback/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

57%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The content is a clear, well-organized recovery procedure with a concrete example, but it pads with obvious explanations, uses template placeholders for tool calls, and leaves error handling implicit rather than an explicit feedback loop.

Suggestions

Trim the 'Why This Works' and 'Tips' sections to non-obvious points only, removing facts Claude already knows.

Make the recovery steps fully executable by giving literal tool-call syntax instead of placeholder templates.

Add an explicit validate-fix-retry loop in Step 3 (e.g. 'If errors appear, fix the script and re-run; only proceed when output is confirmed').

DimensionReasoningScore

Conciseness

The body is mostly efficient, but 'Why This Works' and parts of 'Tips' restate self-evident facts Claude already knows ('python3 in the shell uses the same environment', 'Add explicit print statements for debugging').

2 / 3

Actionability

Concrete commands and a full pandas example are provided, but the recovery steps use template/pseudocode tool-call blocks ('content: <your complete Python code as a string>') rather than literal executable invocations.

2 / 3

Workflow Clarity

A clear four-step sequence is present with a verify step, but the error-recovery checkpoint is implicit ('if any, diagnose and fix the script') rather than an explicit validate-fix-retry loop.

2 / 3

Progressive Disclosure

No bundle files exist and none are referenced; the skill is under 50 lines, single-purpose, and organized into clear labeled sections, meeting the simple-skill allowance for full marks.

3 / 3

Total

9

/

12

Passed

Description

42%

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 conveys a clear, specific niche and an explicit (if narrow) trigger, but its trigger term is internal tool jargon rather than natural user language, limiting discoverability.

Suggestions

Add a natural-language 'Use when...' clause with terms users actually say, e.g. 'Use when Python code execution fails or the sandbox errors out'.

Replace or supplement 'execute_code_sandbox' jargon with user-facing phrasing like 'code execution fails' or 'sandbox errors'.

Broaden the trigger beyond a single tool name to cover general cases (e.g. long-running scripts, file-output tasks that error in the sandbox).

DimensionReasoningScore

Specificity

Names the Python execution domain and concrete actions ('writing to file and running via shell') but describes a single fallback mechanism rather than multiple distinct actions, so it does not reach the comprehensive anchor.

2 / 3

Completeness

It states what the skill does and includes a 'when execute_code_sandbox fails' clause, but the trigger is a single narrow internal-tool condition rather than the broad, natural 'Use when...' guidance that defines the score-3 anchor.

2 / 3

Trigger Term Quality

The only trigger is the internal tool name 'execute_code_sandbox fails', which is technical jargon a user would not naturally say; no common natural-language keywords (e.g. 'Python won't run', 'code execution failed') are present.

1 / 3

Distinctiveness Conflict Risk

The sandbox-failure fallback niche is distinct and tied to a specific tool failure, making it unlikely to trigger for unrelated skills.

3 / 3

Total

8

/

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.

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
HKUDS/OpenSpace
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.