habit-weekly
Skill4 starsChanged 48 days agoMITView on GitHub
Produce a stable personal development-habit weekly report from WezDeck observability: pane-scoped agent concurrency, Claude/Grok/Codex skills+slash+CLI+MCP, CDP verify→iterate, hotkey intensity (Alt+l etc.), optional WakaTime summaries, and optional push to a private habit archive repo. Default window is the previous complete Mon–Sun (not the in-progress week). Use when the user says 开发习惯周报 / 习惯周报 / habit weekly / 这周热键和 agent 习惯 / 出一份习惯数据报告. Prefer this over ad-hoc chat analysis once habit-report exists.
- Reads credentials
In this repository11 other files it gives its agents.
- AGENTS.mdAGENTS.md
- frontend-designSkill
- web-design-guidelinesSkill
- adversarial-reviewSkill
- brainstormSkill
- cross-repo-delegateSkill
- human-runSkill
- repo-hygieneSkill
- wezdeck-runtime-opsSkill
- worktree-recycleSkill
- yuns-engineerSkill
The file
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 0 are distinctive, and 3 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 16
Never: ask the human to paste habit-report.sh as the primary path — load this skill and run the co-located runner.
- Very specific: line 123
Don't 在未要求时省略 --push 却声称已推远端;也不要在仅 --write 时 push
- Very specific: line 124
Don't 把本 skill 链进 ~/.agents/skills / link-platform-skills.sh(仓库内专用)
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Habit weekly(仓库内 skill — 开发习惯周报)
Who runs: the coding agent, not the human.
Never ask the human to paste habit-report.sh as the primary path — load this skill and run the co-located runner.
Scope: WezDeck 仓库内 skill(skills/habit-weekly/)。由 AGENTS.md 路由加载;不经 link-platform-skills.sh 同步到 ~/.agents/skills / ~/.claude/skills(依赖本仓 habit-report.sh,用户级软链会把 repo_root 算错)。
数据收集器是 scripts/dev/habit-report.sh(插件化 Claude / Grok / Codex + 可选 WakaTime)。本 skill 负责:定窗口 → 拉 JSON → 按固定模板渲染周报 →(可选)落盘 →(显式)推送到习惯归档仓。
When to load / run
| 用户意图 | 你做 |
| 开发习惯周报 / 习惯周报 / habit weekly | run.sh(默认上一完整 Mon–Sun;若归档仓落后则 catch-up 接到上次 until+1;默认含 WakaTime) |
| 上周习惯 / last week | run.sh 或 run.sh --week last(与默认相同,含 catch-up) |
| 只要单周、不要接假期缺口 | run.sh --no-catch-up |
| 本周迄今 / this week | run.sh --week this(Mon→today,未完结周;不做 catch-up) |
| 自定义区间 | run.sh --since YYYY-MM-DD --until YYYY-MM-DD(显式窗口,跳过 catch-up) |
| 只要原始 JSON | run.sh --json-only |
| 落盘归档(本机) | run.sh --write(再 --stdout 可同时打印正文) |
| 落盘并推送到私有习惯仓 | run.sh --write --push(--push 隐含 --write) |
Skip / redirect:
- 团队业务周报(commit / TAPD / 飞书)→ coco-weekly-report(不是本 skill)
- 只要某一天的时间线事件流 →
workflow-timeline.sh - 只要热键表、不要 Agent 段 →
habit-report.sh --no-hotkeys的反义词是--providers裁剪;仍可用本 skill 再口头删段
Resolve TOOL_HOME
1. $HABIT_WEEKLY_HOME(显式覆盖)
2. $WEZDECK_REPO/skills/habit-weekly(本机主仓)
3. 本 SKILL.md 同目录(skills/habit-weekly,run.sh 共存)
4. $WEZDECK_ROOT/skills/habit-weeklyTOOL_HOME="${HABIT_WEEKLY_HOME:-$WEZDECK_REPO/skills/habit-weekly}"
R="$TOOL_HOME/run.sh"run.sh 用 pwd -P 解析自身目录,并以 $WEZDECK_REPO(若含 scripts/dev/habit-report.sh)优先定仓库根。不要从 ~/.agents/skills/habit-weekly 启动。
Agent procedure
- 定窗口(用户没说则默认上一完整 Mon–Sun;未完结周须显式
--week this):- 默认 / 上周完整周 → 不传或
--week last - Catch-up(默认开):读归档仓最新
habit-weekly-*_to_<until>.md;若until+1早于默认窗口起点(假期 / 漏跑),把since前移到until+1,until仍为上完整周日——一份报告接缺口,不要再问用户切不切开。--no-catch-up可关;显式--since/--until与--week this不做 catch-up。 - 本周迄今(未完结)→
--week this - 点名日期 →
--since/--until,或--week YYYY-MM-DD(取该日所在周 Mon–Sun)
- 默认 / 上周完整周 → 不传或
- 跑收集 + 渲染(优先
--write落盘;用户要远端归档时再加--push):
仅本机落盘:"$R" --write --push
未完结本周(少用):"$R" --write"$R" --week this --write - 读
references/REPORT_STYLE.md,核对渲染稿是否需补一句人话结论(数字已被模板写好;只在环路归纳与分工上可加 1–2 句,禁止改数字)。窗口若因 catch-up 超过 7 天,结论里可点一句「含假期/空档」。 - 汇报:把 markdown 正文(或落盘路径 + 摘要)用简体中文交给用户。披露数据窗口、
max_running口径(pane+TTL)、WakaTime 是否成功、是否发生了 catch-up。
报告固定结构
见 references/REPORT_STYLE.md。核心段:
- 一句话结论
- WakaTime 时间投入(无 key 则降级说明)
- Agent 并发(pane+TTL;raw_sid 仅诊断)
- Skill / CLI / MCP + 分 provider
- Token / 费用(分 agent · 分模型;Claude
cost-state/ Grokusage.json/ Codexlast_token_usage) - CDP 验证→迭代(有则写)
- 热键调度(Alt+l 为主)
- 工作环路归纳
- 数据边界 + 复现命令
Options
| Flag | 作用 |
--week this|last|DATE | 默认 last(完整 Mon–Sun + catch-up);this=Mon→today;DATE 取其所在周 Mon–Sun |
--no-catch-up | 关闭归档接续;last 只出单周 Mon–Sun |
--since / --until | 显式闭区间(优先于 --week;跳过 catch-up) |
--providers csv | 默认 claude,grok,codex |
--wakatime / --no-wakatime | 默认开;关则跳过 WakaTime |
--write | 写入 $WSL_WORKFLOW_DIR/habit-weekly/habit-weekly-<since>_to_<until>.{md,json} |
--push | 隐含 --write;复制到归档仓 reports/YYYY/ 并 git commit + push(显式,不偷偷推) |
--stdout | 与 --write 联用时额外打印正文 |
--json-only | 只吐 habit-report JSON |
--out-dir DIR | 覆盖落盘目录 |
Archive config
私有习惯仓(非 coco 业务周报):
- 默认远程:
yunsii/wezdeck-habit-weekly - 本机偏好:
~/.config/habit-weekly/state.json→.archive.{repo,local_path,branch,push_json} - 环境覆盖:
HABIT_WEEKLY_ARCHIVE_REPO/HABIT_WEEKLY_ARCHIVE_PATH/HABIT_WEEKLY_ARCHIVE_BRANCH/HABIT_WEEKLY_PUSH_JSON - WakaTime 读统计:
WAKATIME_OAUTH_TOKEN(waka_tok_…,Bearer)优先;写心跳仍用插件侧WAKATIME_API_KEY。落盘:~/.config/shell-env.d/wakatime.env(经runtime_env_load_managed)
Don't
- Don't 把
raw_sid_max_running写成体感并发 - Don't 用含整夜的
avg_running当白天主指标 - Don't 手改 JSON 里的计数再渲染
- Don't 与 coco-weekly-report 混用(一个是习惯可观测,一个是业务交付周报)
- Don't 在未要求时省略
--push却声称已推远端;也不要在仅--write时 push - Don't 把本 skill 链进
~/.agents/skills/link-platform-skills.sh(仓库内专用)
Related
- Collector:
scripts/dev/habit-report.sh+scripts/dev/habit_report/providers/+habit_report/wakatime.py - Push:
skills/habit-weekly/push-archive.sh - Docs:
docs/diagnostics.md→ Habit report - Sibling timeline:
scripts/dev/workflow-timeline.sh
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

