菜单
When the user wants queue progress counts by status, or to audit the queue for inconsistencies (orphans, broken deps, false-done). Do NOT use to modify task status.
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
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。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/nuncaeslupus/ai-chat-exporter --skill queue-status命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
When the user needs claude-arsenal/ set up in a host repo, or wants to register a workspace via --workspace. Re-running is safe (refreshes stale bundle files only). Do NOT use to add tasks (see queue-add) or resume the worker loop (see continue).
Use whenever anything under src/core/parsers/ is being added or repaired — capturing a live DOM, deriving selectors, implementing a BaseParser subclass, and verifying live that every widget the chatbot renders survives into an export. Triggered by "add a parser for X", "support a new chatbot", "capture DOM selectors", "the parser stopped working", "exports come out empty", "deep research is missing from the export". Do NOT use for src/core/exporters/ work (see docs/dev/adding-exporters.md), nor for a parser change that touches no selector.
Use whenever anything under src/core/exporters/ is being added or repaired — a new output format, a content type missing from some formats, or an exported file that will not open. Triggered by "add EPUB/CSV export", "add a new export format", "images are missing from the DOCX", "the exported PDF is blank", "code blocks look wrong in Word". Do NOT use for extracting content off a chat page (see the parser-generator skill), nor for popup or download plumbing that touches no exporter.
基于 SOC 职业分类
正在显示 SKILL.md
| name | queue-status |
| description | When the user wants queue progress counts by status, or to audit the queue for inconsistencies (orphans, broken deps, false-done). Do NOT use to modify task status. |
| user-invocable | true |
| argument-hint | [--detail] |
Reports task counts by status from claude-arsenal/queue/tasks.jsonl: total, open, in-progress, done, merged, and blocked. With --detail, lists each task's ID, title, status, assignee, and unmet dependency IDs.
CANARY: queue-status-loaded-2026-06-13-fb78d23e-d4e5f6a7b8c9d0e1
Load this skill when:
blocked status with unmet deps).# Summary counts
python3 .claude/skills/queue-status/scripts/query_status.py
# Full task list
python3 .claude/skills/queue-status/scripts/query_status.py --detail
The counts above describe progress; for a read-only integrity audit run the bundle's queue doctor — queue_doctor.sh, in claude-arsenal/bin/. It flags orphaned payloads, broken or cyclic dependencies, crashed in_progress claims, stale or branch:-only pr fields, likely secrets committed into payloads, and — when gh is available — false-done (a done/merged row whose PR never merged).
# Audit the queue (auto-enables the gh / git layers when those tools are present)
queue_doctor.sh
# As a CI / make gate: exit non-zero on findings at/above the chosen severity
queue_doctor.sh --fail-on error
It never writes. The orchestrator runs it at session start as an advisory report; run it standalone to gate CI or a make target. See claude-arsenal/AGENTS.md for the session-start wiring.
in_progress tasks with no active assignee signal a crashed session. Use the bundle release script to reset it to open status (see claude-arsenal/AGENTS.md). The consistency check above flags these as stranded-in-progress.blocked does not mean failed. A task becomes eligible automatically once its dependencies complete.