backtest-diagnose

Diagnose failed or underperforming backtests, locate the root cause, and fix the issue

Quality

73%

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 ./a_全网优秀资源/10_大模型/07_skill包/vibe_trading_skills/backtest-diagnose/SKILL.md

The canonical home for this skill is backtest-diagnose in HKUDS/Vibe-Trading

Quality

Content

81%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 concise, actionable, and unusually strong on workflow clarity with explicit validation checkpoints and feedback loops. The main gap is progressive disclosure: everything is inline with no references, which is fine but leaves minor room to split the larger tables out.

Suggestions

Consider moving the Error Taxonomy tables into a references/error-taxonomy.md file and linking to it from the body to improve progressive disclosure.

Make the remaining high-level table fixes more concrete (e.g. replace "Add length checks" with an explicit `if len(df) < N: return pd.Series([])` snippet).

Trim the action_items examples to one canonical example plus the format rule to reduce token weight.

Dimension	Reasoning	Score
Conciseness	The body is lean and well-structured with tables and lists, and it does not explain concepts Claude already knows (e.g. what a DataFrame or backtest is); a few sections (action_items examples, Hard-Gate overlap with the taxonomy) could be trimmed slightly.	4 / 5
Actionability	It gives concrete tool calls (`read_file`, `edit_file`, `bash("pip install xxx")`), a copy-paste AST validation command, and specific file paths, but some table fixes remain high-level ("Add length checks", "Ensure the return value is `pd.Series`").	4 / 5
Workflow Clarity	The Diagnostic Workflow is an explicit 5-step sequence with a verification checkpoint, plus a Hard-Gate Checklist, a rerun-and-limit-to-3-iterations feedback loop, and Post-Fix Validation Rules — matching the anchor for clear sequence with explicit validation and error-recovery loops.	5 / 5
Progressive Disclosure	The skill is well-organized with clear section headers and no nested references (no bundle files exist), but at ~80 inline lines the error-taxonomy tables and action_items rules could arguably live in separate reference files; structure is good with minor organization gaps.	4 / 5
	Total	17 / 20 Passed

Description

66%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 states concrete actions and uses natural trigger terms, but it omits any explicit "when to use" guidance, which caps completeness. Adding a trigger clause would move it firmly into the top band.

Suggestions

Append an explicit trigger clause, e.g. "Use when a backtest fails, raises an error, or produces poor/abnormal results."

Add a couple of natural synonyms users might say (e.g. "poor results", "errored", "strategy underperformance") to broaden trigger coverage.

Tighten the verbs with a more specific fix action (e.g. "patch signal_engine.py and rerun") to push specificity toward 5.

Dimension	Reasoning	Score
Specificity	"Diagnose failed or underperforming backtests, locate the root cause, and fix the issue" lists three concrete actions (diagnose, locate root cause, fix) tied to a named domain, but the verbs stay fairly high-level compared to the anchor listing granular operations.	4 / 5
Completeness	It clearly states what the skill does but has no "Use when..." clause or equivalent explicit trigger guidance, so per the judging guideline completeness is capped at 3.	3 / 5
Trigger Term Quality	"failed", "underperforming", and "backtests" are natural phrases a user would say when they need this skill, giving good coverage, though a few common synonyms (e.g. "poor results", "errored") are absent.	4 / 5
Distinctiveness Conflict Risk	"backtest" diagnosis is a clear niche distinct from most skills, but without an explicit trigger clause the description could still overlap with general debugging or code-repair skills.	4 / 5
	Total	15 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: charliedream1/ai_quant_trade
Path: a_全网优秀资源/10_大模型/07_skill包/vibe_trading_skills/backtest-diagnose/SKILL.md
Commit: 6565e03

Reviewed: 3 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.