trading-wisdom
Core trading insights learned from Agent Arena competition. Use when making any trading decision to apply institutional knowledge.
25
16%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./data/skills-md/0xhubed/agent-trading-arena/trading-wisdom/SKILL.mdQuality
Discovery
32%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This description is vague and lacks concrete actions, relying on buzzwords like 'institutional knowledge' and 'core trading insights' without specifying what the skill actually does. While it includes a 'Use when' clause, the trigger is overly broad ('any trading decision'), making it indistinguishable from other potential trading skills. The description needs substantial improvement in specificity and distinctiveness.
Suggestions
Replace 'core trading insights' with specific concrete actions the skill performs, e.g., 'Applies position sizing rules, risk/reward thresholds, and entry/exit criteria learned from Agent Arena competition.'
Narrow the 'Use when' clause to specific scenarios instead of 'any trading decision', e.g., 'Use when evaluating trade setups, sizing positions, setting stop losses, or reviewing market conditions.'
Add natural trigger terms users would actually say, such as 'trade', 'buy', 'sell', 'position size', 'stop loss', 'risk management', 'entry point'.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description uses vague language like 'core trading insights' and 'institutional knowledge' without listing any concrete actions. It does not specify what the skill actually does (e.g., analyze positions, calculate risk, generate signals). | 1 / 3 |
Completeness | The 'what' is extremely weak ('core trading insights' is not a concrete capability), but there is a 'when' clause ('Use when making any trading decision'). However, the 'when' is overly broad and the 'what' is essentially missing, so it barely reaches a 2. | 2 / 3 |
Trigger Term Quality | It includes 'trading decision' which is a natural keyword, but lacks specific variations users might say such as 'buy', 'sell', 'position sizing', 'risk management', 'market analysis', or 'portfolio'. 'Agent Arena competition' is very niche jargon. | 2 / 3 |
Distinctiveness Conflict Risk | 'Any trading decision' is extremely broad and would conflict with any other trading-related skill. There are no distinct triggers that would help Claude differentiate this skill from other trading or finance skills. | 1 / 3 |
Total | 6 / 12 Passed |
Implementation
0%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill is an auto-generated data dump of trading competition observations rather than an actionable skill document. It suffers from extreme verbosity with massive redundancy (the same insights about zero-trade strategies and overtrading are repeated 5-10 times each), lacks any executable guidance or decision workflow, and presents all content as a flat monolithic list. The content would benefit enormously from distillation into a concise decision framework with clear steps.
Suggestions
Distill the 30+ winning strategies and 45+ anti-patterns into 5-7 core principles with a clear decision tree (e.g., 'Step 1: Assess market regime → Step 2: If moderate bull, reduce trade frequency to <10/day → Step 3: If bearish, prefer zero-trade')
Remove redundant entries that express the same insight - e.g., consolidate all 'zero-trade preserves capital' variants into one entry with supporting data, and all 'overtrading loses money' variants into another
Add a concrete workflow section with specific decision criteria: market regime thresholds, trade frequency targets per regime, entry/exit rules, and validation checkpoints before executing trades
Move the detailed pattern catalog to a separate reference file (e.g., PATTERNS.md) and keep only the top 5-7 highest-confidence actionable rules in the main SKILL.md
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Extremely verbose at 400+ lines with massive redundancy. Many patterns repeat the same insight (zero-trade preserves capital, overtrading loses money) with slight variations. The 'Patterns to Avoid' section has ~45 entries, many saying essentially the same thing. This could be condensed to perhaps 20% of its current size without losing information. | 1 / 3 |
Actionability | The content describes observations and historical patterns but provides no concrete, executable guidance. There are no code examples, no specific commands, no decision trees or algorithms to follow. It reads as a post-mortem analysis rather than actionable instructions for making trading decisions. | 1 / 3 |
Workflow Clarity | There is no clear workflow or decision-making process defined. The skill lacks any sequenced steps like 'assess market regime → determine trade frequency → select assets → validate entry.' It's a flat list of observations with no structured process for applying them to actual trading decisions. | 1 / 3 |
Progressive Disclosure | The content is a monolithic wall of text with no references to external files and no hierarchical organization. All 206 patterns are dumped inline. The winning strategies and patterns to avoid sections contain enormous repetitive detail that should be summarized with references to detailed data elsewhere. | 1 / 3 |
Total | 4 / 12 Passed |
Validation
100%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 11 / 11 Passed
Validation for skill structure
No warnings or errors.
- Repository
- NeverSight/skills_feed
- Commit
25d5403
- 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.