laozhong86

Enables autonomous issue delivery in gxpm by persisting user authorization for seamless workflow management.

Identifies architectural improvements in codebases, enhancing testability and AI-navigability through structured refactoring suggestions.

开启 gxpm 自动驾驶模式。使用 Autopilot Grant 持久化用户授权，让代理在授权 profile 内自主完成 issue delivery。

MUST use during the qa phase before transitioning to land. Headless browser automation for QA evidence capture. Use when user asks to test a web page, take a screenshot, verify an element, fill a form, or capture browser evidence for an issue.

Compile and type-check verification. Use after implementing code changes, before committing, or when the build fails and root cause is unclear.

MUST use during the cleanup phase before transitioning to ship. 多 issue worktree 合并前的代码清理与简化。在 cleanup 阶段识别跨 issue 重复、命名不一致、接口错位和死代码。

Systematic issue debugging using graph-powered code navigation. Use when user asks to trace a bug, investigate a specific error, or find the root cause of a failing test or exception.

Disciplined diagnosis loop for hard bugs and performance regressions. Use when user says 'diagnose this', reports a hard bug, describes a performance regression, or asks why something fails.

Skill quality evaluation harness for static analysis. Use when adding a new skill, modifying skill structure, auditing skill quality, or checking for governance compliance.

Navigate and understand codebase structure using the knowledge graph. Use when user asks how code works, wants to understand architecture, trace execution flows, or explore unfamiliar parts of the codebase.

Create feedback issues in the gxpm source repository when the agent discovers problems or improvement opportunities during gxpm workflow execution. Use when the agent notices a gxpm bug, process flaw, skill deficiency, or has a suggestion for the gxpm tool itself.

Grilling session that challenges plans against the domain model and sharpens terminology. Use when user wants to stress-test a plan, align on requirements, challenge assumptions, or make architecture decisions.

Compact the current session into a handoff document so another agent can continue the work. Use when switching agents, ending a session, or passing work between implementer subagents.

Pre-commit hygiene and atomic commit discipline. Use before every commit, or when preparing changes for self-review or ship.

Subagent implementer behavior template enforcing four-dimension self-review. Use when dispatching a task to an implementer subagent, or when a subagent needs structured self-review and reporting rules.

Linear CLI integration for gxpm issue lifecycle management. Use when creating, updating, or querying Linear issues, syncing gxpm issues to Linear, or managing issue lifecycle via command line.

MUST use during the plan phase to write implementation-plan before transitioning to dispatch. Break plans into independently-grabbable vertical-slice issues. Use when converting a plan into tickets, breaking a feature down, or turning discussion into a structured PRD.

Build a throwaway prototype to validate a design before committing to it. Use when you need to sanity-check a data model, state machine, or UI design before writing behavior-spec.

Plan and execute safe refactoring using dependency analysis. Use when user asks to rename, extract, split, move, or simplify code, or when a code review suggests refactoring.

Agent Army 并行审查模式：Stage 1 spec compliance gate 通过后，Stage 2 才并行扇出 code-quality / security / test / accessibility reviewers。仿 superpowers subagent-driven-development 双阶段模板。