Now liveThe Skillselion MCP - thousands of ranked skills, loaded into your agent mid-task. No install.Get it →
jongwony avatar

Jongwony Epistemic Protocols

  • 125 repo stars
  • Updated July 28, 2026
  • jongwony/epistemic-protocols

jongwony-epistemic-protocols is a Claude Code plugin that structures human–AI interaction quality at every decision point via thirteen epistemic protocols.

About

jongwony-epistemic-protocols is a Claude Code plugin pack that structures how you and your agent know things together at every decision point—not generic prompting tips, but named protocols for intent, framing, comprehension, validation, and reflection. developers install it when agent output sounds confident but underspecified, when scope creeps because assumptions were never surfaced, or when cross-session work loses shared ground. The bundle ships thirteen plugins spanning inquiry, bounding, cooperative interpretation, risk surfacing, and execution-time checks, aligned with the public handbook at epistemic-protocols.com. Because the invoke-when is essentially any meaningful human–AI choice, journeyScope is journey-wide: Idea research, Validate scope, Build planning, Ship review, and Grow content all benefit. It is methodology, not a single API integration—think philosopher-grade guardrails for dialogue quality rather than a deploy script. Intermediate complexity reflects learning protocol names and when to route to each; confidence is high given explicit domain branding and rich keyword catalog in the repo metadata.

  • 13-protocol Claude Code bundle for human–AI interaction quality
  • Decision-point rituals: intent, framing, comprehension, validation, and reflection
  • Cross-session and quest-based patterns for sustained agent collaboration
  • Handbook-style coverage: dialogue, boundaries, multiperspective analysis, execution grounding
  • Documented at epistemic-protocols.com with dashboard and onboarding motifs

Jongwony Epistemic Protocols by the numbers

  • Data as of Jul 28, 2026 (Skillselion catalog sync)
/plugin install jongwony-epistemic-protocols@jongwony/epistemic-protocols

Add your badge

Show developers this plugin is listed on Skillselion. Paste this into your README.

Listed on Skillselion
repo stars125
Last updatedJuly 28, 2026
Repositoryjongwony/epistemic-protocols

What it does

Install structured epistemic protocols so every human–AI decision in Claude Code has clearer intent, boundaries, and mutual understanding.

Who is it for?

Reflective developers who want repeatable interaction quality—not just output volume—across planning, building, shipping, and iteration.

Skip if: Skip if you only want fast codegen with no overhead for framing, dialogue discipline, or epistemic checkpoints.

What you get

After registration, you can route Claude Code sessions through explicit protocols so intent, comprehension, and validation are structurally clearer before you act.

  • Thirteen registered protocol plugins in Claude Code
  • Repeatable decision-point workflows for intent, bounds, and validation
  • Clearer session reports and cross-session grounding patterns

By the numbers

  • 13 plugins in bundle
  • Repository: jongwony/epistemic-protocols
  • Public site: epistemic-protocols.com

Recommended Plugins

How it compares

Journey-wide epistemic methodology pack (13 plugins), not a one-shot integration skill or MCP tool.

FAQ

Who is Jongwony Epistemic Protocols for?

Claude Code users who care about decision quality, shared understanding, and structured human–AI dialogue from research through execution.

When should I use Jongwony Epistemic Protocols?

Use at decision points whenever intent is vague, risks are implicit, or you need validation before trusting agent plans, code, or content.

How do I add Jongwony Epistemic Protocols to my agent?

Install the jongwony/epistemic-protocols plugin bundle in Claude Code and follow epistemic-protocols.com onboarding to pick protocols for your session.

AI Agentsagentsresearch

This week in AI coding

Five minutes, every Monday - the tools, releases and tactics for developers.

unsubscribe anytime.