verify-success-after-unknown-error

Verify task completion by checking filesystem state when execute_code_sandbox or run_shell return misleading unknown errors

Quality

71%

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/verify-success-after-unknown-error/SKILL.md

Quality

Content

67%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 a well-structured, actionable verification workflow with explicit decision logic and a feedback loop. Its main weaknesses are redundant verification code across sections, assumed tool API signatures, and an underspecified partial-success recovery branch.

Suggestions

Consolidate the repeated file-existence checks into a single canonical snippet and have later sections reference it, to reduce redundancy and improve conciseness.

Replace the assumed 'from tools import list_dir, read_file' API with the real tool call signatures, or clearly mark them as placeholders, so the Python examples are genuinely executable.

Expand the 'partial success' branch of the decision logic with a concrete recovery procedure rather than the vague 'may need targeted fix'.

Dimension	Reasoning	Score
Conciseness	The body avoids basic-concept padding but repeats the same file-existence verification logic across Step 1, Step 2, the 'Complete Pattern', and the shell example, which is more than minor redundancy and could be tightened.	3 / 5
Actionability	Concrete Python and bash code plus a clear decision tree cover the common cases, but the examples assume unverified tool APIs ('from tools import list_dir, read_file' and guessed signatures), leaving minor gaps versus fully executable code.	4 / 5
Workflow Clarity	A clear three-step sequence with content validation and an explicit decision checkpoint plus a proceed/retry feedback loop is present; the underspecified 'partial success -> targeted fix' branch is a minor validation gap keeping it below a 5.	4 / 5
Progressive Disclosure	Sections are well organized with no nested references and all content appropriately inline (nothing clearly belongs in a separate file), but at roughly 130 lines the under-50-line simple-skill exception to reach 5 does not apply.	4 / 5
	Total	15 / 20 Passed

Description

75%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 is specific, distinct, and cleanly answers both what the skill does and when to use it via an explicit trigger clause. Its main weakness is trigger-term coverage, which depends on jargon tool names rather than natural synonyms.

Dimension	Reasoning	Score
Specificity	Names the domain plus one to two concrete actions ('Verify task completion', 'checking filesystem state'), matching the anchor for a clear domain with limited actions rather than the 'several specific actions' of a 4.	3 / 5
Completeness	It explicitly states the what ('Verify task completion by checking filesystem state') and an explicit when clause with concrete triggers ('when execute_code_sandbox or run_shell return misleading unknown errors'), satisfying the anchor for clearly answering both.	5 / 5
Trigger Term Quality	'unknown errors' is a natural trigger phrase, but coverage leans on jargon tool names (execute_code_sandbox, run_shell) with no synonyms or common variations, fitting the 'some relevant keywords but missing variations' anchor.	3 / 5
Distinctiveness Conflict Risk	The trigger is a very specific niche (named tools plus a particular error condition) with distinct triggers and minimal overlap risk, matching the 'clear niche with distinct triggers' anchor.	5 / 5
	Total	16 / 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/verify-success-after-unknown-error/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.