Construct the strongest coherent research mainline from the current evidence pack.
Treat the evidence pack as a derived PI snapshot; current facts are owned by
findings/results, frontier/incubator state, committed Gems state, and generation
boundaries.
Use validation signals as follow-up leads, not as clean promotion facts.
Fixed review questions:
What is the strongest mechanism explanation right now?
Which conjecture family is most worth continuing to exploit?
Which clarity or counterexample-coverage arm has the most remaining headroom?
Which findings can be synthesized into a cross-peer hypothesis?
Which next-generation peer should advance the mainline?
Use private KB only as disclosed support, and keep every claim traceable to evidence cards or prior findings.
1---2name: builder-pi-23description: Toy Math Builder PI4---5# Toy Math Builder PI67Construct the strongest coherent research mainline from the current evidence pack.8Treat the evidence pack as a derived PI snapshot; current facts are owned by9findings/results, frontier/incubator state, committed Gems state, and generation10boundaries.11Use validation signals as follow-up leads, not as clean promotion facts.1213Fixed review questions:14- What is the strongest mechanism explanation right now?15- Which conjecture family is most worth continuing to exploit?16- Which clarity or counterexample-coverage arm has the most remaining headroom?17- Which findings can be synthesized into a cross-peer hypothesis?18- Which next-generation peer should advance the mainline?1920Use private KB only as disclosed support, and keep every claim traceable to evidence cards or prior findings.
Run npx skillmds@latest add sapientinc/builder-pi-2 in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.
Toy Math Builder PI It is listed under Coding & Dev Tools on SkillMD.
This skill has not completed SkillMD's automated safety review yet. Capability flags: docs only. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.
This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.
Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.
sapientinc (@sapientinc) published this skill. Their other Agent Skills are listed on their SkillMD profile.
sapientinc