Skip to content
SkillScript-led1,215 tokens in use

Edge Candidate Agent

Detects edge ideas in end-of-day price data, writes research tickets, and exports validated strategy.yaml and metadata.json specs for a backtest pipeline.

7,426 stars on the repository MIT Licensetradermonty/claude-trading-skills · skills/edge-candidate-agent

Skill label

measured from its files

What installing Edge Candidate Agent costs

85tokens idle
1,215tokens in use
2,597tokens, up to

Idle is what sits in context while the skill is installed. In use is its instructions, loaded when it runs. Up to adds the files its instructions link. Counts are approximate, made offline with cl100k_base.

Model dependence Script-led

Script-ledMixedModel-led

Most of this skill is scripts, so it behaves much the same whichever model runs it.

This describes the skill's structure, how much of it is scripts. It does not say how well it works. Scripts: scripts/auto_detect_candidates.py, scripts/candidate_contract.py, scripts/export_candidate.py, scripts/validate_candidate.py.

Touches

shelldetectednetworkdetectedwrites filesdetected

Declared tags come from the skill's frontmatter; detected ones from its scripts, its code blocks and the commands it names in inline code. This is detected, not audited.

How it scores

Quality

How well the skill is written, counted from its own files: its frontmatter, its steps, examples, stated output and constraints, its size and whether every file it ships is one it names.How we score it →
Excellent
  • Frontmatter12/15
  • Structure23/25
  • Examples12/15
  • Output15/15
  • Size15/15
  • Resources9/15

Security

Starts at 100 and loses points for each thing in the skill's files that a reader should look at first.How we score it →
Good
  • Runs shell commands−5

    does this on the machine, and the frontmatter does not say so

    SKILL.md, line 70
  • Reaches the network−10

    does this on the machine, and the frontmatter does not say so

    scripts/auto_detect_candidates.py, line 611
  • Writes files−8

    does this on the machine, and the frontmatter does not say so

    scripts/auto_detect_candidates.py, line 1517

Repository health

Pre-computed score based on development activity, maintenance, community, maturity, and trend momentum.How we score it →
Good
  • Development activity100
  • Maintenance52
  • Community68
  • Maturity36
  • Momentum40

Edge Candidate Agent turns end-of-day OHLCV observations, anomalies and optional hints into reproducible research tickets for US equity long-side ideas. It then exports the exportable ones as a strategy.yaml and a metadata.json that match the edge-finder-candidate/v1 interface of a separate trade-strategy-pipeline repository.

Three Python scripts detect candidates, export them and validate them against the interface and the pipeline’s schema rules. In the split edge workflow it handles the final export and validation stage. It prepares research specs and gives no investment advice.

What You Get

  • A strategies/<candidate_id>/ folder with strategy.yaml and metadata.json for each exportable ticket
  • Daily detection files: daily_report.md, market_summary.json, anomalies.json and watchlist.csv
  • Exportable and research-only ticket YAML files
  • A pass or fail result with reasons from the interface validator

Common Use Cases

  • Running a daily scan of end-of-day data to list new candidate ideas
  • Converting a written research ticket into files a backtest pipeline accepts
  • Checking a candidate against the v1 interface before a dry run

What Makes It Unique

It is built around a fixed interface contract and refuses candidates that break it: schema bounds, a folder name that differs from the id, empty conditions and missing interface metadata. The export rules limit v1 to two supported entry families and mark everything else research-only. It pairs with four sibling skills that cover the earlier stages.

Under The Hood

The agent runs Python scripts from the skill folder. Auto-detection reads a local OHLCV file and an optional hints file, and writes report and ticket files into an output directory. The export script writes candidate folders into the strategies directory of another repository. When you pass a pipeline root, the validator imports that repository’s schema code, and the agent may run uv there. An optional external command can supply ideas. The skill needs PyYAML and the target pipeline checkout.

Join founders building with open source

Opinionated takes, migration guides, cost-saving tips, and insights from the open source ecosystem.

Subscribe on SubstackJoin 750+ subscribers