shell-python-fallback

Use run_shell with embedded Python heredoc as reliable fallback when code execution tools fail

Quality

73%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Medium

Suggest reviewing before use

Fix and improve this skill with Tessl

tessl review fix ./benchmarks/gdpval/skills/shell-python-fallback/SKILL.md

Quality

Content

72%

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

The content is highly actionable with executable examples and clean section organization, but it repeats the core pattern multiple times and lacks an integrated validation/retry loop in its main workflow.

Suggestions

Collapse the redundant heredoc demonstrations into one canonical example plus one realistic end-to-end example to reduce token cost.

Turn step 5 into an explicit feedback loop (e.g. 'if stdout shows an error, apply the matching Troubleshooting row and re-run') rather than deferring recovery to a separate table.

Trim or remove the 'Why This Works' section, which restates sandbox-bypass behavior Claude can infer from the pattern.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient and assumes Claude knows Python, but the heredoc pattern is demonstrated redundantly (basic pattern, step 3, and two full examples) and the 'Why This Works' section re-explains the run_shell/sandbox distinction, so it could be tightened.	2 / 3
Actionability	It provides fully executable, copy-paste-ready code (reportlab PDF generation, CSV-to-JSON processing), concrete step-by-step instructions, and specific best-practice and troubleshooting guidance.	3 / 3
Workflow Clarity	A clear five-step sequence is present with a verification step (step 5), but the validation checkpoint is light and error recovery is offloaded to a separate troubleshooting table rather than an inline validate-fix-retry feedback loop.	2 / 3
Progressive Disclosure	The single-purpose skill is well-organized into clearly headed sections (When to Use, The Pattern, Step-by-Step, Examples, Best Practices, Troubleshooting) with no external references needed and no nested-reference problems, giving easy navigation.	3 / 3
	Total	10 / 12 Passed

Description

75%

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 concrete and covers both what and when with an explicit trigger, but it is built around a single action and uses technical jargon over the natural terms a user would say when hitting this failure mode.

Suggestions

Add natural-language trigger terms a user would actually say (e.g. 'when Python execution returns unknown errors' or 'when the code sandbox keeps failing') alongside the technical phrasing.

Consider listing a second concrete capability (e.g. 'run multi-line Python scripts reliably') to broaden the specificity beyond a single composite action.

Dimension	Reasoning	Score
Specificity	It names a concrete tool and technique ('run_shell with embedded Python heredoc') but describes a single composite action rather than enumerating multiple distinct capabilities, so it does not reach the 'multiple specific concrete actions' anchor.	2 / 3
Completeness	It explicitly states both what it does ('Use run_shell with embedded Python heredoc as reliable fallback') and when to use it ('when code execution tools fail'), satisfying the explicit-trigger requirement that would otherwise cap completeness at 2.	3 / 3
Trigger Term Quality	It includes a relevant trigger phrase ('when code execution tools fail') but leans on technical jargon ('run_shell', 'Python heredoc') and omits natural variations a user would actually say, such as 'unknown error' or 'Python won't run', which appear only in the body.	2 / 3
Distinctiveness Conflict Risk	The narrow niche of a shell-heredoc fallback for failed code-execution tools, paired with a specific remedy, makes it unlikely to trigger for the wrong skill.	3 / 3
	Total	10 / 12 Passed

Validation

100%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: HKUDS/OpenSpace
Path: benchmarks/gdpval/skills/shell-python-fallback/SKILL.md
Commit: c5a9c4b

Reviewed: 2 days ago

Table of Contents

Discovery Implementation Validation

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.