CP

cmoralesm/p123-claude-skill

Developer tools

이 페이지는 AI로 자동 번역되었습니다. 기술적인 세부 사항은 원문을 확인하세요. 원문 보기

22 stars 품질 68 트렌드 19MIT

Agent Skill for Portfolio123 (P123): 464 functions, 4456 factors, verified line-by-line against official docs. Factor investing, screening, ranking systems, backtesting, and p123api Python wrapper.

개요

Carlos Morales runs QuantSolvings, a boutique quantitative practice in factor investing for equities: coaching, consulting and team training on Portfolio123. Book a free 45-minute introductory call. A Claude Skill that turns Claude into a reliable Portfolio123 (P123) assistant: writing screen rules and formulas, building ranking-system XML, replicating published factors, and pulling data through the REST API with the official p123api Python wrapper. is the current release. 4.1.0 was the first public release since v3.0.0 and carried two releases' worth of work; 4.1.1 adds live verification of the regional universe IDs, the last item still open from the v3.0.0 community reports. - a factor-replication reference and an API refresh: - (new): 35 recipes across value, quality, earnings quality, momentum, low risk, size and investment/issuance, each with the P123 implementation, the ranking node's direction and scope, and the translation traps.

README

원문 README의 일부입니다. 원문 보기.

Portfolio123 Claude Skill

Built by a Verified Portfolio123 Coach and Consultant. Carlos Morales runs QuantSolvings, a boutique quantitative practice in factor investing for equities: coaching, consulting and team training on Portfolio123. Book a free 45-minute introductory call.

A Claude Skill that turns Claude into a reliable Portfolio123 (P123) assistant: writing screen rules and formulas, building ranking-system XML, replicating published factors, and pulling data through the REST API with the official p123api Python wrapper.

v4.1.1 is the current release. 4.1.0 was the first public release since v3.0.0 and carried two releases’ worth of work; 4.1.1 adds live verification of the regional universe IDs, the last item still open from the v3.0.0 community reports.

New in 4.1 - a factor-replication reference and an API refresh:

  • references/factor-replication.md (new): 35 recipes across value, quality, earnings quality, momentum, low risk, size and investment/issuance, each with the P123 implementation, the ranking node’s direction and scope, and the translation traps. No performance figures appear anywhere in it, by design: an unverifiable alpha is an invented fact. Seven constructions P123 genuinely cannot reproduce are listed as such, with the closest honest proxy instead of a fake.
  • The REST API surface is current again. P123 added six operations on 2026-08-28 - GET /dataSeries, GET /rank, POST /rank/create, GET /strategy, POST /strategy/{id}/copy and POST /strategy/{id}/copy-book - plus the preproc machine-learning block on POST /data/universe and an extended currency enum. references/api.md now documents all 32 paths / 39 operations of the 2026-09-03 spec capture.
  • The GET /dataSeries correction. v4.0.0 stated that P123 had no Data Series read operation and used that absence to explain a reported 405. That is no longer true, and the file says so in those words rather than quietly deleting the passage.
  • p123api 3.1.0, verified method by method: all 44 public client methods, the nine that now return typed objects instead of dicts, the four keyword-only lookups, and a 2.x-to-3.x migration table.

From 4.0.0 (finished but never published) - four defects found by a P123 practitioner using the skill on live research work: the ^ power operator is now the documented idiom for powers and

GitHub에서 전체 README 보기

추천 도구

다른 키워드를 입력하거나 필터를 제거해 보세요.

설치

npx skillfish add cmoralesm/p123-claude-skill
MCP 초보자라면? mcpskill.org 가이드부터 시작하세요 →