菜单
Human-in-the-loop macOS debugging checkpoints. Use when automation needs a human to confirm UI state, report observations, or proceed through manual steps.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Brainstorming specialist. Use before implementing any feature, fix or cli design to discuss the approach with the user first. Plans data models, storage layouts, test scenarios, and expected outputs.
Produce LOOP_<YYYY-MM-DD>_<slug>.md — a real runnable repro/verify runbook from user logs or prose. Expands to precise CLI/URL ops, builds workspace tools, runs steps 1–4 for real, records SYMPTOM CONFIRMED or VERIFY PASS. Pair with run-the-loop for fixes. Use for /establish-a-loop, "establish a loop", or "create a debug loop".
Fix until done: goal + inspect script under script/debug/, then iterate implement → inspect until PASS. Use after establish-a-loop or standalone for /run-the-loop, "loop until done", "max effort", "don't stop until".
Clarification-phase followup. Use when the user mentions followup to stay in discussion mode without moving to implementation.
OpenAI-compatible HTTP mock server for testing agents without real LLM calls. Loads a JSON config of request→response exchanges and returns pre-configured responses. Use when writing doctests or integration tests for agent runners.
Incremental deduplication workflow: map overlapping implementations, plan a PR stack into shared packages, seal doctests, migrate callers without behavior change. Use when the user mentions consolidate, dedupe, shared package, move enrichment into dot-pkgs, eliminate duplication, or /consolidate-code. Pair with investigate (overlap map) and doctest-tdd (sealed tests → impl).
| name | debug-with-user |
| description | Human-in-the-loop macOS debugging checkpoints. Use when automation needs a human to confirm UI state, report observations, or proceed through manual steps. |
Human-assisted debugging for macOS workflows. Automate everything scriptable; at checkpoints, ask the user via native dialogs and parse structured JSON answers.
/debug-with-user or explicit human-assisted debug requestsdebug-with-user ask with a clear title, message, and absolute paths.via == "button": use affirmed (true/false) for pass/fail semantics.via == "free_text": treat answer as the user's report (Customize path).0 answer, 1 dismissed/cancel, 2 error (e.g. non-macOS without dry-run).debug-with-user ask \
--title "Step 1 — Did VS Code open?" \
--message "Project folder:\n/tmp/demo" \
--option "Yes — window opened" \
--option "No — window did not open" \
--affirm "Yes — window opened" \
--cancel "Cancel"
The CLI always appends a Customize button for free-text reports.
macOS display alert allows at most 3 buttons. The CLI keeps Customize and
up to 3 options total (2 --option plus Customize). More than that exits
with an error suggesting multi-step confirmation (no silent truncation). Every preset returns a normal answer
(affirmed true/false); press Escape to dismiss without choosing.
Aliases: confirm (--yes / --no), proceed (--proceed / --cancel).
If debug-with-user is not on PATH:
go install github.com/xhd2015/agent-pro/agents/debug-with-user@latest
Real dialogs require macOS osascript. In CI, set DEBUG_WITH_USER_DRY_RUN=1 and stage
responses with DEBUG_WITH_USER_DRY_RUN_BUTTON, DEBUG_WITH_USER_DRY_RUN_TEXT, or
DEBUG_WITH_USER_DRY_RUN_DISMISSED=1.
Preset button:
{"answer":"Yes — window opened","via":"button","affirmed":true}
Customize (free text):
{"answer":"VS Code opened but wrong workspace","via":"free_text"}