strategy-generate

Create, modify, and optimize quantitative trading strategies, then backtest and evaluate them.

Quality

79%

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 ./agent/src/skills/strategy-generate/SKILL.md

Quality

Content

85%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 dense, highly actionable operating manual: complete config and contract examples, exact commands, a validated 7-step workflow with feedback loops and hard gates. Its weaknesses are a dangling examples.md reference with no actual bundle files, config documentation that belongs in a separate reference file, and minor verbose passages.

Suggestions

Create the referenced examples.md (or remove the link) — the Supporting Files section points to a file that does not exist in the bundle.

Move the ~60 lines of config.json parameter documentation (source, interval, optimizer, validation keys) into a references/config.md and keep only the filled example plus the top 3-4 critical keys inline.

Tighten the rebalance_tolerance and warm-up prose into rule-style bullets; the multi-sentence justifications ("0.05 is a reasonable starting point, not a recommendation with evidence behind it...") can be halved without losing the guidance.

Dimension	Reasoning	Score
Conciseness	Nearly every section carries engine-specific, non-inferable detail (warmup_bars semantics, position_adjustment modes, OKX pair formats, PIT-merge rules) with no padding about concepts Claude already knows. Not 5 because a few passages run long — the hedging prose around "0.05 is a reasonable starting point, not a recommendation with evidence behind it" and the warm-up failure narrative could be tightened, and "Cryptocurrency Notes" partly duplicates the Market Detection table.	4 / 5
Actionability	Fully executable guidance: a complete copy-paste config.json, an exact syntax-check bash command, the full SignalEngine contract with docstring, exact regexes per market, exact code formats ("BTC-USDT", "AAPL.US"), and concrete action_items examples. This matches the anchor "copy-paste ready code or commands; specific examples cover the common cases".	5 / 5
Workflow Clarity	The 7-step workflow is clearly sequenced with an explicit validation checkpoint (step 4 syntax check), a feedback loop (step 7: edit_file → backtest → re-evaluate), a self-check quality checklist, and hard gates for evaluation. This matches the top anchor: explicit validation steps, error-recovery loops, and checklists.	5 / 5
Progressive Disclosure	The bundle contains no references/, scripts/, or assets/ directories, yet the body links to [examples.md](examples.md) as "example call sequence" — a dangling reference — plus a cross-skill link. Meanwhile ~60 lines of config.json parameter documentation are inlined in SKILL.md rather than split into a reference file. Structure is well-organized (not anchor 2's "minimal structure"), but the broken reference and inlined bulk fit anchor 3.	3 / 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 clearly states what the skill does with concrete domain-specific actions, but it lacks any "Use when..." trigger guidance, which caps completeness and limits its discoverability. Keyword coverage is good though missing common synonyms like "algo trading" or "quant".

Suggestions

Add an explicit trigger clause, e.g. "Use when the user asks to backtest, optimize, or evaluate a trading strategy, or mentions quant/algo trading, Sharpe ratio, or portfolio performance."

Include natural synonyms users actually say — "algorithmic trading", "algo trading", "quant strategy", "portfolio backtest" — to broaden trigger coverage.

Mention the concrete deliverables (writes signal_engine.py and config.json) to sharpen the what-portion.

Dimension	Reasoning	Score
Specificity	Lists several concrete actions — "Create, modify, and optimize quantitative trading strategies, then backtest and evaluate them" — naming the domain and 5 distinct operations. Not 5 because it omits concrete scope signals (markets, instruments, outputs like config/signal code); not 3 because coverage goes well beyond 1-2 actions.	4 / 5
Completeness	The "what" is clear (create/modify/optimize strategies, backtest and evaluate), but there is no "Use when..." clause or equivalent trigger guidance — the rubric explicitly caps this at 3. It fits anchor 3 exactly: clear what, when entirely missing.	3 / 5
Trigger Term Quality	"quantitative trading strategies" and "backtest" are natural phrases users would say when needing this skill. Not 5 because common synonyms like "algo/algorithmic trading", "quant", "portfolio", or "Sharpe" are absent; not 3 because the present terms go beyond merely "some relevant keywords".	4 / 5
Distinctiveness Conflict Risk	"Backtest" and "quantitative trading strategies" carve a distinct niche unlikely to fire for unrelated skills. Not 5 because without trigger phrases it could still overlap with generic coding or data-analysis requests; not 3 because the domain is far from generic.	4 / 5
	Total	15 / 20 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
relative_links	Relative link issues: 1 missing, 1 suspicious	Warning

	Total	14 / 16 Passed

Repository: HKUDS/Vibe-Trading
Path: agent/src/skills/strategy-generate/SKILL.md
Commit: 251b094

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.