edge-candidate-agent

Generate and prioritize US equity long-side edge research tickets from EOD observations, then export pipeline-ready candidate specs for trade-strategy-pipeline Phase I. Use when users ask to turn hypotheses/anomalies into reproducible research tickets, convert validated ideas into `strategy.yaml` + `metadata.json`, or preflight-check interface compatibility (`edge-finder-candidate/v1`) before running pipeline backtests.

Quality

86%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

87%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

Well-structured, actionable, and token-efficient content with good progressive disclosure. The main gap is an explicit validation feedback loop in the workflow for what is effectively a batch export-and-handoff operation.

Suggestions

Add an explicit validation checkpoint in the Workflow: after step 5, state 'If validation fails: review the reported reasons, fix the ticket/strategy, and re-run validate_candidate.py before proceeding.'

Make the dry-run step a numbered checkpoint with a concrete command (e.g. 'Run --dry-run first; only proceed to full pipeline execution on pass') rather than only a Guardrail bullet.

Clarify the error-recovery path when export_candidate.py rejects a ticket (e.g. move to research-only vs. fix and re-export) so the feedback loop is unambiguous.

Dimension	Reasoning	Score
Conciseness	The body is lean: each section (Overview, When to Use, Output, Workflow, Quick Commands, Export Rules, Guardrails, Resources) earns its place and avoids explaining concepts Claude already knows; no padding around YAML, OHLCV, or pipeline basics.	3 / 3
Actionability	Quick Commands provides fully executable, copy-paste-ready bash invocations with real flags and paths, and the workflow plus Export Rules/Guardrails give concrete, specific directives rather than abstract descriptions.	3 / 3
Workflow Clarity	The 6-step workflow is sequenced and validation is present (validate_candidate.py at step 5, a --dry-run guardrail), but the validate->fix->retry feedback loop is implicit rather than an explicit checkpoint with error-recovery instructions.	2 / 3
Progressive Disclosure	A concise overview points to real one-level-deep references and scripts (all referenced paths verified to exist in references/ and scripts/), with a Resources section giving one-line navigation for each file.	3 / 3
	Total	11 / 12 Passed

Description

85%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong, specific description with explicit what/when structure and a clear niche. Its only weakness is trigger-term naturalness, which relies on domain jargon rather than plain user phrasings.

Suggestions

Soften jargon in the 'Use when' clause by adding plain-language triggers (e.g. 'when users want to turn trading ideas into backtest-ready specs') alongside the existing technical phrasings.

Reduce reliance on version/interface identifiers like 'edge-finder-candidate/v1' in the trigger clause; keep them for the what-statement but use everyday terms for when-to-use.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'Generate and prioritize US equity long-side edge research tickets', 'export pipeline-ready candidate specs', 'convert validated ideas into strategy.yaml + metadata.json', and 'preflight-check interface compatibility' — each tied to specific artifacts.	3 / 3
Completeness	Explicitly answers both what (generate/prioritize research tickets, export candidate specs) and when via a clear 'Use when users ask to...' clause listing three concrete triggers.	3 / 3
Trigger Term Quality	Includes several relevant trigger phrasings ('turn hypotheses/anomalies into reproducible research tickets', 'preflight-check interface compatibility') but leans heavily on niche jargon ('edge-finder-candidate/v1', 'EOD observations', 'trade-strategy-pipeline Phase I') rather than common variations a user would naturally say.	2 / 3
Distinctiveness Conflict Risk	The niche is sharply scoped to US equity long-side edge research and a named pipeline interface, with distinct triggers unlikely to fire for unrelated skills.	3 / 3
	Total	11 / 12 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: tradermonty/claude-trading-skills
Path: skills/edge-candidate-agent/SKILL.md
Commit: 62a1635

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.