fallback-doc-with-delegation

Generate professional documents after tool failures using shell_agent for complex formats or write_file for simple text/markdown

Quality

61%

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/write-file-fallback-report-enhanced/SKILL.md

Quality

Content

60%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 body is well-structured and actionable with concrete code and a clear step sequence, including validation and success criteria. Its main weakness is significant redundancy: the core method-selection decision and the report template are repeated multiple times across sections.

Suggestions

Eliminate the repeated report template and the repeated shell_agent-vs-write_file decision; keep one authoritative decision section (table or tree) and reference it from the Best Practices, Method Comparison, and Decision Checklist sections.

Add an explicit validate-and-retry feedback loop in Step 6 for when shell_agent or write_file fails, rather than only confirming the file exists.

Replace [placeholder] tokens in at least one end-to-end example with realistic concrete values so the guidance is fully copy-paste executable.

Dimension	Reasoning	Score
Conciseness	The shell_agent-vs-write_file decision is restated five or more times (Step 2 table, decision tree, Code Example, Method Comparison table, Decision Checklist) and the markdown report template appears verbatim in Step 3B, Step 4, and the Code Example, making the body noticeably padded with redundant sections.	2 / 5
Actionability	It provides concrete shell_agent() and write_file() calls, a worked SBAR PDF example, a generate_fallback_document function, and list_dir validation, all of which are mostly executable, though several snippets rely on [placeholder] tokens rather than being fully copy-paste ready.	4 / 5
Workflow Clarity	Steps 1 through 6 are clearly sequenced with an explicit decision point at Step 1, a validation step at Step 6, and a Success Criteria checklist, but there is no validate-then-fix feedback loop for the case where generation fails.	4 / 5
Progressive Disclosure	The skill is self-contained with no bundle files and is organized into well-labeled sections and tables, but at ~250 lines it is longer than the simple-skill threshold and contains inlined content (comparison tables, checklists) that could be trimmed or split.	4 / 5
	Total	14 / 20 Passed

Description

62%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 clearly communicates a fallback document-generation niche tied to a concrete trigger condition and names the two delegation methods. It is somewhat held back by tool-jargon trigger terms and a single composite action rather than a richer action list.

Suggestions

Add natural user-facing trigger terms (e.g. "reports", "PDF", "create a document") alongside the tool names so the description matches phrases users actually say.

Tighten the "when" clause to specify which failures (e.g. "when read_file, search_web, or execute_code_sandbox fail") for a more explicit trigger.

Expand the action list beyond the single "generate" verb to distinct capabilities (e.g. "generate, format, and save documents") to lift specificity.

Dimension	Reasoning	Score
Specificity	"Generate professional documents" names the domain and pairs it with two concrete methods ("shell_agent for complex formats" and "write_file for simple text/markdown"), but the action set is essentially one task split by tool rather than a comprehensive list of distinct actions.	3 / 5
Completeness	It states both what (generate professional documents via shell_agent/write_file) and when ("after tool failures"), which is an explicit trigger clause, but the when could be more specific about which failures or scenarios.	4 / 5
Trigger Term Quality	It surfaces relevant keywords ("documents", "tool failures", "markdown", "text") but leans on technical tool names (shell_agent, write_file) and misses common natural variations a user would say like "report", "PDF", or "create a file".	3 / 5
Distinctiveness Conflict Risk	The "after tool failures" fallback niche is a fairly distinct trigger that is unlikely to fire for ordinary document creation, though the broad "generate professional documents" framing keeps some minor overlap risk with general writing skills.	4 / 5
	Total	14 / 20 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/write-file-fallback-report-enhanced/SKILL.md
Commit: 3827781

Reviewed: 5 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.