pdf-gen-fallback-dee20c

Fallback workflow for document generation when code sandbox returns opaque errors

Quality

72%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./benchmarks/gdpval/skills/pdf-gen-fallback-dee20c/SKILL.md

Quality

Content

77%

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

The skill body is highly actionable with executable code and a clear, validated workflow, but it is held back by repeated example code and the absence of any progressive-disclosure structure splitting the inline content.

Suggestions

De-duplicate the reportlab Canvas example: show the full error-handling version once and reference it from the complete-workflow example instead of repeating it.

Move the troubleshooting table and best practices into a reference file (e.g. references/troubleshooting.md) and link to it, improving progressive disclosure for this >50-line skill.

Tighten the fallback hierarchy section so each library lists the concrete run_shell command rather than only a one-line description.

Dimension	Reasoning	Score
Conciseness	The body avoids explaining basic concepts Claude knows and is well-sectioned, but the reportlab Canvas generation code is repeated three times (step 2, step 3, and the complete example), which is redundant padding that could be tightened, so it sits at 2 rather than the lean bar of 3.	2 / 3
Actionability	It provides fully executable `run_shell: python -c` snippets with real reportlab code, library availability checks, pip install, and file verification commands that are copy-paste ready, meeting the score-3 bar.	3 / 3
Workflow Clarity	A clear five-step sequence (check library, execute, handle errors, verify output, fallback hierarchy) with explicit validation checkpoints (library check, try/except with exit codes, ls/file verification) and a troubleshooting feedback table, satisfying the score-3 anchor; the operation is file generation rather than destructive/batch so no cap applies.	3 / 3
Progressive Disclosure	The body is well organized into clear sections, but at over 50 lines with no bundle files it is a single monolithic file with all content inline and no one-level-deep external references, so it does not reach the score-3 split-and-navigate bar.	2 / 3
	Total	10 / 12 Passed

Description

67%

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 pairs a what with an explicit when trigger, but its action list is thin and its trigger terms lean on internal jargon ("code sandbox", "opaque errors") rather than phrasings users naturally say.

Suggestions

Add concrete actions to raise specificity, e.g. "Generate PDFs and reports via run_shell when execute_code_sandbox returns opaque errors."

Replace jargon with natural user phrasings: "PDF or report generation fails with an unclear error message" instead of "code sandbox returns opaque errors".

Broaden trigger coverage to include terms users actually say: PDF, report, document, failed, error message, couldn't generate.

Dimension	Reasoning	Score
Specificity	Names the domain ("document generation") and a single action area ("Fallback workflow"), but does not list multiple specific concrete actions like generate PDFs, build reports, or handle errors, so it stops at score 2 rather than reaching the multi-action bar of 3.	2 / 3
Completeness	It explicitly answers both what ("Fallback workflow for document generation") and when ("when code sandbox returns opaque errors") with an explicit trigger clause, satisfying the score-3 bar and avoiding the missing-trigger cap at 2.	3 / 3
Trigger Term Quality	"document generation" is a reasonably natural phrase, but "code sandbox" and "opaque errors" are internal/technical jargon users would rarely say, and common variations like PDF, report, failed, or unclear error message are missing, so it lands at 2 rather than the broad natural-term coverage of 3.	2 / 3
Distinctiveness Conflict Risk	The fallback-on-sandbox-errors trigger is a fairly distinct niche, but the broad "document generation" domain could still overlap with a primary document/PDF generation skill, so it is somewhat specific rather than clearly unconflicting at 3.	2 / 3
	Total	9 / 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/pdf-gen-fallback-dee20c/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.