li-gong-demo

Simulates a backend engineer persona for architecture reviews, release workflows, and team alignment training.

Updated Aug 15, 2026
One-click install
npx skills add https://github.com/qqlcx5/skills-hub --skill li-gong-demo-qqlcx5
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: li-gong-demo
Source: https://github.com/qqlcx5/skills-hub/tree/main/skills/immortal-skill/examples/li-gong-demo
Command: npx skills add https://github.com/qqlcx5/skills-hub --skill li-gong-demo-qqlcx5

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Teams often need a consistent reference for how a specific senior backend engineer approaches architecture reviews, release procedures, and cross-team communication, but that knowledge is scattered across chat logs, MR comments, and checklists. This Skill consolidates that evidence into a reusable persona simulation for internal alignment and training. ## Core Features & Use Cases - Evidence-grounded persona simulation: Reproduces Li Gong's communication style, questioning patterns, and decision criteria, with every behavior tagged by source type (verbatim, artifact, impression). - Procedural knowledge base: Encodes Go backend workflows including mini-RFC writing, MR review standards, canary release stages (1% → 10% → 50% → 100%), and monitoring requirements. - Conflict resolution records: Documents known contradictions (e.g., 30-minute vs 10-minute canary observation windows) with temporary rulings. - Use Case: During onboarding, a new team member asks the persona how to prepare a release; the Skill responds with Li Gong's actual checklist requirements, escalation boundaries, and typical pushback phrasing. ## Quick Start Ask the agent to respond as Li Gong when reviewing my release plan for a Go microservice deployment.

Frequently Asked Questions about li-gong-demo

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I simulate a colleague's working style with an AI agent?

Provide the agent with structured persona files covering interaction style, procedural knowledge, and known conflicts, each claim tagged by evidence source. This Skill demonstrates the pattern using chat records, MR comments, and release checklists from a backend engineer.

What files does a persona simulation skill need?

This Skill uses SKILL.md for activation rules, interaction.md for communication style, procedure.md for workflows and standards, and conflicts.md for unresolved contradictions. A manifest.json records sources and file fingerprints.

Can this persona make commitments on behalf of the real person?

No. The Skill explicitly forbids fabricating external commitments attributable to the real person and requires stating uncertainty when source material is missing. It is scoped to internal alignment and training only.

What are the limitations of persona simulation based on chat records?

Entries tagged as impression are subjective collaborator opinions, not self-reported facts. Domains with insufficient source material, such as non-backend work for this persona, are explicitly out of simulation scope.

How are conflicting statements in source material handled?

Conflicts are recorded in conflicts.md with both claims, their sources and dates, plus a temporary ruling. For example, 30-minute versus 10-minute canary observation windows are kept in parallel with scenario-based rulings.