athenara:~$ registry man skills/tradermonty-trading-skills
Claude Trading Skills (tradermonty)
A collection of 72 Claude skills for individual equity traders, chained into 11 workflow manifests covering market-regime checks, screening, position sizing, risk gates, journaling and postmortems.
added 2026-08-17 · MIT · external
SYNOPSIS
$ git clone https://github.com/tradermonty/claude-trading-skillsTARGETS
claude-code, claude-app
DESCRIPTION
The repository ships exactly 72 skills, each with its own SKILL.md under skills/<name>/, plus
72 prebuilt .skill archives under skill-packages/ for upload into the Claude app. A canonical
skills-index.yaml catalogues all of them across eight categories: market-regime (16),
strategy-research (14), swing-opportunity (7), trade-planning (7), advanced-satellite (7),
core-portfolio (6), trade-memory (6) and meta (9). Eleven machine-readable manifests under
workflows/ chain skills into routines such as market-regime-daily, core-portfolio-weekly,
swing-opportunity-daily, trade-memory-loop and monthly-performance-review. Installation is a copy:
drop a skills/<name>/ folder into ~/.claude/skills/ for personal use or .claude/skills/
inside a project, or upload the matching .skill ZIP under Customize > Skills in the Claude app.
A documented no-paid-API starter path runs five skills on public CSVs and local files — market-breadth-analyzer, uptrend-analyzer, position-sizer (pure calculation, no I/O), trader-memory-core (local YAML journaling) and signal-postmortem. Beyond that, optional paid integrations are itemised per skill in a README table: Financial Modeling Prep (free tier of 250 requests/day), FINVIZ Elite ($39.50/month or $299.50/year) and Alpaca (free paper trading, via an MCP server); many skills need none. The project states its own limits plainly — “This is not a signal service or a promise of profitability”, it “is not designed for fully automated trading, signal outsourcing, or short-term scalping”, and past performance, backtests, screens and AI-generated analysis do not guarantee future results.
Skill quality is self-declared, not demonstrated. The index labels 58 skills “production” and 14
“beta”, but its per-skill verification block covers eight dimensions — instruction contract, unit
tests, workflow contract, end-to-end replay, data provenance, financial logic review, empirical
validation and security review — and across all 72 skills every field reads either not_verified
(383) or not_applicable (81). No skill has any dimension marked verified. The author is
pseudonymous: the GitHub handle carries no real name, bio or company, and the LICENSE copyright
line reads “TraderMonty”; that LICENSE file is unambiguous MIT, while the README’s own License
section only says the material is “provided for educational and research purposes”. Several skills
are meta-tooling (skill-designer, skill-idea-miner, trading-skills-navigator) or personal
workflows rather than trading, so the effective trading-skill count is below 72. Development is
active — last commit 2026-08-12, with 2,649 stars, 615 forks and 24+ contributors as of
2026-08-17.
PREREQUISITES
trading [●●●··] moderate ai [●●···] basic programming [●●···] basic setup [●●···] basic
METADATA