Reconstruct AI programming conversations and project delivery materials into a verifiable project evidence chain, helping users clarify problems, decisions, individual actions, delivery phases, and actual impacts before deciding whether to work on job application expression or uplevel practice.
适用范围
Scope of Application
仅处理以下材料：
人工智能编程对话；
项目设计、实现、联调、验收或交付记录；
用于说明项目是否落地、效果如何和个人负责边界的材料。
普通求职经历提升、岗位定位和招聘开场白交给
/great-resume
；本入口不修改简历，
也不启动其他学习状态。
Only processes the following materials:
AI programming conversations;
Project design, implementation, integration, acceptance, or delivery records;
Materials used to demonstrate whether a project is implemented, its effectiveness, and the scope of individual responsibility.
General job experience enhancement, role positioning, and recruitment opening lines are handled by
/great-resume
; this entry does not modify resumes, nor does it initiate other learning states.
Use user decisions, verifiable results, and key assistant reasoning from the entire conversation as materials, and do not treat speculations as facts.
Conduct a privacy check before output: project codenames, tokens, keys, emails, customer identifiers, internal paths, and undisclosed architectures are generalized or replaced by default.
First determine the delivery phase, then organize the nature of the evidence. Only use "production deployment", "internal pilot", "prototype/technical verification", or "planning/estimation" when explicitly stated in the materials.
Output according to the nine-stage evidence chain in order, without using polished wording to cover up evidence gaps.
If the user submits their own rewriting attempt afterwards, only provide feedback on fact retention, uplevel actions, and cross-boundary risks, and do not rewrite the full recap on their behalf.
Problem Background: Real scenarios, original problems, constraints, and desired outcomes.
Solution Decision: What solution was adopted, why it was adopted, what was abandoned, and key trade-offs.
Individual Actions: Design, implementation, integration, verification, or promotion actions actually completed by the user; parts completed by the team, platform, and AI are marked separately.
Delivery Status: Explicitly marked as production deployment, internal pilot, prototype/technical verification, or planning/estimation.
Implementation Scope: Environment, user or team scope, duration, covered processes, and technical scope.
Effectiveness Evidence: Itemize measured results, phased results, technical verification, or estimated benefits, and maintain the original range of indicators.
Individual Boundary: Explain the parts the user is responsible for, participates in, integrates, verifies, and is not responsible for; do not upgrade team or platform achievements to individual leadership.
Pending Evidence: List missing user, scope, time, indicator, acceptance, or usage records, and do not infer to fill in as facts.
Interview Follow-up Questions: Generate follow-up questions that can be answered by existing materials, focusing on solution trade-offs, failure paths, implementation scope, effectiveness evidence, and responsibility boundaries.
表达边界
Expression Boundaries
已落地、试点、原型和计划必须显式区分。
实测结果、阶段结果、技术验证和估算收益必须显式区分。
没有可靠数字时使用可核验的定性表达，不编造比例、用户量、成本或排名。
不输出原始密钥、令牌、邮箱、客户标识、内部代号和未公开实现细节。
不把演示原型写成业务上线，不把估算收益写成已经实现的结果。
机制命名必须能回到当前材料中的真实行为、控制流或边界。
Implemented, pilot, prototype, and planning must be explicitly distinguished.
Measured results, phased results, technical verification, and estimated benefits must be explicitly distinguished.
When there are no reliable figures, use verifiable qualitative expressions, and do not fabricate proportions, user volumes, costs, or rankings.
Do not output original keys, tokens, emails, customer identifiers, internal codenames, or undisclosed implementation details.
Do not present demo prototypes as business launches, nor present estimated benefits as achieved results.
Mechanism naming must correspond to real behaviors, control flows, or boundaries in the current materials.
After the nine-stage recap is completed, Uplevel Migration is included by default. Only when the user explicitly requests job application, resume, or packaging expression can Candidate Resume Statements be added. The rewriting exercises in Uplevel Migration are optional, and users do not need to continue practicing.
Uplevel Migration includes at most:
One sentence of original facts;
One high-density expression connecting facts, mechanisms, and impacts;
The uplevel actions used this time;
One optional rewriting exercise based on the same material.
Judgment criteria for uplevel actions:
High Density: One sentence connects facts, mechanisms, and impacts at the same time;
High Impact: Highlights key decisions, causal levers, and individual actions without relying on exaggerated adjectives;
High Dimensionality: Covers technical mechanisms, delivery phases, evidence nature, individual boundaries, and user or business value simultaneously.
用户提交练习后的反馈
Feedback After User Submits Practice
只回答三点：
哪些事实被正确保留；
哪些升维动作已经成立；
哪些表达存在证据缺口或个人边界越界。
不强制继续练习，不创建跨对话学习状态。
Only answer three points:
Which facts are correctly retained;
Which uplevel actions are already valid;
Which expressions have evidence gaps or cross individual boundaries.
Do not force continued practice, nor create cross-conversation learning states.
发送前检查
Pre-Send Checklist
九段证据链是否完整且顺序正确；
交付阶段和效果证据是否分别标注；
个人动作是否和团队、平台、人工智能的动作分开；
缺失信息是否进入待补证据；
是否完成隐私泛化；
升维迁移是否默认加入；
候选简历表述是否只在用户提出相应需求后出现。
Is the nine-stage evidence chain complete and in the correct order;
Are delivery phases and effectiveness evidence marked separately;
Are individual actions separated from those of the team, platform, and AI;
Is missing information included in pending evidence;
Is privacy generalization completed;
Is Uplevel Migration included by default;
Does Candidate Resume Statements only appear after the user makes a corresponding request.