🐛 Debugging

Every AI output has structural blind spots determined by the generation process itself. Future Tokens operations are named, composable instruments that target specific blind spots. They surface new information on every pass because expanding the output changes the blind spot geometry.

Interactive system flow tracing across CODE, API, AUTH, DATA, NETWORK layers with SQLite persistence and Mermaid export. Use for security audits, compliance documentation, flow tracing, feature ideation, brainstorming, debugging, architecture reviews, or incident post-mortems. Triggers on audit, trace flow, document flow, security review, debug flow, brainstorm, architecture review, post-mortem, incident review.

Detect and fix AI-generated writing patterns (slop). Comprehensive detection with 45+ patterns, tiered severity scoring, and editor mode.

>

>

Comprehensive application log investigation workflow that reads logs end-to-end, correlates signals with code paths and runtime context, and identifies evidence-backed issues with impact and remediation steps. Use when users ask to analyze logs, investigate incidents, find root causes from log records, explain recurring warnings/errors, or check whether logs reveal hidden system problems.

Read code to identify edge cases, write tests that cover those edge cases, and fix the implementation if tests expose failures. Use when a user asks to add edge-case tests, harden behavior, or validate that existing code handles unusual inputs or error paths (e.g., empty data, nulls, boundaries, invalid formats, timeouts, retries).

Modular security expert workflow for evidence-first red-team audits with two execution modes: default interaction mode and user-requested auto mode. Use when users ask to find exploitable vulnerabilities, prove exploitability with reproducible evidence, implement minimal safe fixes, and add failing-then-passing security regression tests. Includes modules for agent systems and financial programs.

系統化除錯流程：根據使用者描述推定最可能的錯誤原因，對相關程式碼加入額外日誌，提供使用者重現步驟並請其回報日誌，依日誌修復並反覆確認直到問題解決，最後移除所有額外日誌。用於需要迭代式除錯與使用者回報日誌的情境。

Record a tmux proof session demonstrating completed work. Use when the user asks to prove, demonstrate, or show that code actually works.

Use when writing ANY implementation code, fixing bugs, or modifying existing code. Delegates research to a fast subagent that distills WebSearch results into compact RAG-efficient summaries. Main context never sees raw output.

Test and evaluate skill projects during development. Use when the user wants to test, evaluate, or assess a skill that is currently being developed in the workspace — i.e., the skill project in the current directory or a specified skill folder. This skill generates test cases, executes them against the target skill, and produces an evaluation report with strengths, weaknesses, and optimization suggestions. Triggers on requests like "test this skill", "evaluate my skill", "run skill tests", "assess skill quality", or "check if my skill works".

用于根据用户提供的Jenkins Job构建链接进行失败问题诊断的技能。当用户需要分析Jenkins流水线构建失败原因时使用，通过调用Jenkins API获取构建结果、分析失败阶段、检查构建日志和测试报告，提供详细的失败诊断报告。

>

>

Interact with a web browser. Can start a browser, connect to it, evaluate JavaScript, make screenshots, read console logs and let the user select DOM elements. Use when interacting with unknown websites (e.g. scraping or Userscripts) or debugging browser-stuff. Requires uv.