CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

quant-alpha-mining

Use when the user asks to mine controlled alpha hypotheses using idea, implementer and evaluator agents under preregistered research gates.

Quality

80%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./.agents/skills/quant-alpha-mining/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

86%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 an exemplary lean, operational skill: concrete CLI commands, a gated five-step workflow with a preview checkpoint and anti-fabrication reporting rules, and proper deferral of shared details to a single external guide. The only improvements are incremental — an inline YAML skeleton and explicit failure-handling branches would make it fully copy-paste ready.

DimensionReasoningScore

Conciseness

The body is lean and operational throughout: no concept explanations, each sentence adds project-specific constraints ('Do not fabricate AR/IC/drawdown, skip missing PIT evidence, treat unknown usage as zero'), and every token carries guidance Claude could not infer. It assumes Claude's competence and matches anchor 5.

5 / 5

Actionability

Concrete, executable commands are given ('python egs/skills.py preview --config <task.yaml>', 'run --config <task.yaml> --format jsonl', 'providers --probe', 'show/events/files/cancel') plus a pointer to a real example YAML. Minor gaps keep it below anchor 5: no inline task-YAML template or field list, and 'providers --probe' is given without its full command prefix.

4 / 5

Workflow Clarity

Five numbered steps form a clear sequence (read example → create YAML → preview → run once → report actual results), with a pre-run checkpoint (preview, probe when backend availability is uncertain) and error-recovery guidance ('retain failed-run evidence instead of repeating paid calls'). It falls short of anchor 5 because there is no explicit if-fails-then-fix loop for the preview step and no checklist for the reporting step.

4 / 5

Progressive Disclosure

The body is a short, well-organized overview that keeps provider limits, data fields, and research gates in a single clearly signaled one-level-deep reference ([the shared execution guide](../../../docs/usage/agent_cli_skills.md)). No bundle files exist to inline, content is appropriately split, and navigation is easy — matching the simple-skill case for a top score.

5 / 5

Total

18

/

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 is highly distinctive and explicit about when to use the skill, but its capability coverage is thin: it names a single action with an architecture sketch and only one narrow trigger phrase. Adding a few more concrete actions (e.g., backtest evaluation, evidence gating) and natural synonyms (quant, factor, signal) would raise both specificity and trigger quality.

Suggestions

List 2-3 concrete capabilities in the description (e.g., generating candidate factors, running gated backtests, producing evidence-based evaluation reports) instead of the single 'mine controlled alpha hypotheses' action.

Broaden trigger terms with natural synonyms users would say, such as 'quant research', 'factor mining', 'alpha discovery', or 'hypothesis backtesting', not just the exact phrase 'mine controlled alpha hypotheses'.

Break the circular 'when' clause by describing the situations that call for the skill (e.g., 'Use when the user wants preregistered, evidence-gated quant hypothesis generation rather than ad-hoc backtesting') in addition to the literal task phrasing.

DimensionReasoningScore

Specificity

The description names the domain ('mine controlled alpha hypotheses') and a mechanism ('idea, implementer and evaluator agents under preregistered research gates'), which is one concrete action plus architecture detail rather than several distinct actions. It is more specific than anchor 2's generic actions but does not list the multiple concrete operations that anchor 4 requires.

3 / 5

Completeness

Both parts are explicitly present: the 'what' ('mine controlled alpha hypotheses using idea, implementer and evaluator agents under preregistered research gates') and an explicit 'when' ('Use when the user asks to mine controlled alpha hypotheses'). The 'when' clause is explicit but single and circular — it restates the task rather than offering varied concrete trigger phrases, so it falls short of anchor 5.

4 / 5

Trigger Term Quality

Relevant keywords like 'mine', 'alpha hypotheses', 'agents', and 'research gates' are present, but common variations a user would naturally say — 'quant', 'factor mining', 'backtest', 'signal' — are absent. Coverage is relevant yet narrow, matching anchor 3 rather than the broader coverage of anchor 4.

3 / 5

Distinctiveness Conflict Risk

The description occupies a clear niche — preregistered-gated multi-agent alpha mining — with trigger terms unlikely to appear for any other skill. Conflict risk with generic or even adjacent quant skills is minimal, matching anchor 5.

5 / 5

Total

15

/

20

Passed

Validation

93%

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

CriteriaDescriptionResult

relative_links

Relative link issues: 1 suspicious

Warning

Total

15

/

16

Passed

Repository

charliedream1/ai_quant_trade

Path

.agents/skills/quant-alpha-mining/SKILL.md

Reviewed

Table of Contents

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.