CtrlK
BlogDocsLog inGet started
Tessl Logo

code-exec-fallback

Fallback pattern for executing Python code when execute_code_sandbox fails

54

Quality

60%

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/code-exec-fallback/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

61%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 content is well-structured and provides a concrete executable example, making it reasonably actionable and navigable. Its main gaps are cross-section redundancy that hurts conciseness and a total absence of validation/error-recovery steps in a workflow that executes arbitrary shell code.

Suggestions

Add a validation checkpoint after execution, e.g., 'Check run_shell exit code and stderr; if non-zero, fix the script and re-run before proceeding.'

Collapse the redundant guidance between "The Pattern", "Step-by-Step Instructions", and "Tips for Success" into a single concise sequence.

Provide a concrete, copy-paste tool-call form for the steps (write_file/run_shell with real argument syntax) instead of text-block descriptions.

DimensionReasoningScore

Conciseness

The body is mostly efficient but repeats itself across "The Pattern", the step-by-step block, and "Tips for Success" (e.g., timeout and import guidance each appear twice), matching the anchor for mostly-efficient-but-could-be-tightened. Not a 2 because nothing is severely padded or explains concepts Claude doesn't know; not a 4 because the cross-section redundancy is clearly trimmable.

3 / 5

Actionability

The Example gives real, runnable Python plus concrete commands ("python3 analyze.py", timeout=60) and concrete tool calls (write_file/run_shell), matching mostly-executable guidance with minor gaps. Not a 5 because the step-by-step instructions are text-block descriptions ("Use write_file to save...") rather than copy-paste executable form; not a 3 because the worked example supplies genuinely executable code.

4 / 5

Workflow Clarity

A clear 3-step sequence exists (write script -> execute via shell -> handle output/cleanup), but there are no validation checkpoints and no feedback loop for when the fallback itself fails, matching the anchor for steps-listed-but-validation-gaps. Not a 4 because checkpoints are entirely implicit and the skill (shell execution of arbitrary code) carries real risk; the validation-gap cap holds it at 3.

3 / 5

Progressive Disclosure

The skill is self-contained (no references/, scripts/, or assets/ bundle exists) with well-organized sections (When to Use, The Pattern, Steps, Example, Tips, When This Helps), matching good-structure-with-minor-organization-gaps. Not a 5 because the section redundancy (Tips/When This Helps overlapping earlier sections) is a minor organization gap; not a 3 because structure and navigation are clear.

4 / 5

Total

14

/

20

Passed

Description

58%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 cleanly states what the skill does and a concrete trigger condition, giving it strong completeness and distinctiveness. Its weakness is trigger-term quality: the trigger leans on an internal tool name rather than natural user language, and it lists only one action.

Suggestions

Rewrite the trigger in natural user/Claude language, e.g., 'Use when execute_code_sandbox repeatedly fails with timeouts, missing dependencies, or sandbox restrictions.'

Add a second concrete action to broaden capability coverage, e.g., '...write Python to a file and run it via shell, capturing structured output.'

Include common synonyms such as 'code execution', 'sandbox failure', or 'run Python' so the description matches varied phrasings.

DimensionReasoningScore

Specificity

Quotes "Fallback pattern for executing Python code" — it names the domain and one concrete action (executing Python code) but lists no further actions, matching the anchor that names the domain with 1-2 concrete actions yet is not comprehensive. Not a 2 because the action is concrete rather than generic, and not a 4 because only one action is enumerated.

3 / 5

Completeness

It states both the what ("Fallback pattern for executing Python code") and an explicit when ("when execute_code_sandbox fails"), so it is not capped at 3. Not a 5 because the trigger is technical jargon rather than concrete natural trigger phrases; not a 3 because the when clause is explicit rather than weakly implied.

4 / 5

Trigger Term Quality

The only trigger phrase is "when execute_code_sandbox fails", an internal tool name a user would not naturally say; the natural phrases (e.g., "Python won't run", "code keeps failing") are missing. This matches the anchor of one or two technical/generic keywords missing the natural phrases users say; not a 3 because no common variations or synonyms are present.

2 / 5

Distinctiveness Conflict Risk

The trigger "when execute_code_sandbox fails" carves a narrow niche (a fallback for a specific failure mode) with minimal overlap risk against general code-execution skills. Matches the clear-niche/distinct-triggers anchor; not a 4 because the failure-condition trigger is uniquely specific to this skill.

5 / 5

Total

14

/

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
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.