AGENT SKILL

routines

Put the project's recurring rituals on a schedule, with the owner's consent. Weekly north-star report, monthly market watch, dependency triage, Supabase advisor sweep, reflect and harness-optimize. Uses Claude Code Routines (cloud cron) where they fit, and documents the Codex / local cron equivalent. Use when asked to automate, schedule or "keep the project moving on its own", or after setup finishes.

License 许可
MIT
最近更新
2026-10-06
Stars
★ 0
Forks
0

01 前言 Front Matter

name
routines
description
Put the project's recurring rituals on a schedule, with the owner's consent. Weekly north-star report, monthly market watch, dependency triage, Supabase advisor sweep, reflect and harness-optimize. Uses Claude Code Routines (cloud cron) where they fit, and documents the Codex / local cron equivalent. Use when asked to automate, schedule or "keep the project moving on its own", or after setup finishes.
disable-model-invocation
true

03 SKILL.md 正文

Routines

Without a schedule, the rituals only run when the session-start line nags. This skill sets them up to run on their own. It creates nothing without an explicit yes for each routine, because every run spends tokens and can open PRs/issues.

1. Offer the menu (ask, structured, multi-select)

Ritual Cadence What a run produces Where it can run
north-star-report Mon 08:xx Report file + proposed tickets (issue comment; files nothing unapproved) Cloud (needs the PostHog connector on the Routine)
dependency-triage Mon 08:xx triage pass over open dependency PRs + npm audit / pip-audit; one summary issue Cloud
market-watch 1st Tue of the month, 09:xx Re-checks competitors and complaint themes against docs/product/VALIDATION.md; one issue only if something changed Cloud
supabase-advisor Wed 08:xx Security/performance advisors → one issue per new finding Cloud (needs the Supabase connector)
harness-optimize Fri 16:xx Harness-change PR from the week's session signal Local: it reads session transcripts on the machine where you work
reflect Fri 16:xx Memory notes PR Local, same reason

Estimated cost: each cloud run is one agent session, roughly a normal working session's tokens. Say that in the question. Recommend the first two by default.

2. Create the ones they picked

Each ritual's procedure lives in .agents/routines/<ritual>.md (cadence, where it runs, connectors, steps). The Routine's prompt just points there, so the procedure is versioned and reviewed with the code.

Claude Code (cloud Routines). Run /schedule (alias /routines) and create one routine per pick against this GitHub repo, with a prompt that stands alone because each run starts fresh:

In <owner>/<repo>: follow .agents/routines/<ritual>.md end to end.
Never merge, never push to main, never file a ticket the procedure says needs
approval: put proposals in one issue titled "<ritual>: <date>" instead.

Add only the connectors that ritual needs (PostHog, Supabase, Linear). Use a minute that isn't :00 to avoid the top-of-hour rush. Manage or delete them later at claude.ai/code/routines.

Codex / any agent (local cron). Codex has no cloud scheduler; use the machine's:

# crontab -e   (runs in the repo; logs to ~/.local/state/<slug>-rituals.log)
13 8 * * 1  codex exec -C /path/to/repo --approve-for-me -c sandbox_workspace_write.network_access=true "Follow .agents/routines/north-star-report.md." >> ~/.local/state/<slug>-rituals.log 2>&1
47 16 * * 5 codex exec -C /path/to/repo --approve-for-me -c sandbox_workspace_write.network_access=true "Follow .agents/routines/reflect.md, then .agents/routines/harness-optimize.md." >> ~/.local/state/<slug>-rituals.log 2>&1

For Claude Code the local equivalent is cd /path/to/repo && claude -p "/harness-optimize" --permission-mode auto. Cron runs can't answer an approval prompt, so pick a mode that won't stop to ask. On macOS, prefer a launchd agent (cron may lack disk access). These flags were checked against codex-cli 0.159.3 (-C, --approve-for-me, -c); confirm with codex exec --help / claude --help first, since CLIs drift.

3. Record it

Add each scheduled ritual's name to .claude/harness/manifest.json → cadence.scheduled (e.g. ["north-star-report", "dependency-triage"]), so the session-start nudge stops asking you to run by hand what a routine already runs. Commit that on a branch with a PR (chore: schedule rituals), and log it in docs/decision-log.md (who agreed, what runs, where).

Reply with a table: ritual → where it runs → first run time → how to stop it.

Ask the owner

Follow .agents/rules/product-judgement.md: ask with a structured question (recommended option first), never decide these silently. In this skill that means: Which rituals run unattended and on what schedule, since each run spends tokens and can open issues or PRs. One yes per routine.

04 同仓库更多 Skills

foxinthehenhouse
foxinthehenhouse/app-in-a-box

ship

Release checklist for getting a build to testers or the stores. Runs the risk gate first (refuses while a risk item is open) and walks the owner through the compliance checklist. Covers the version bump, EAS build and submit, OTA updates and when they're safe, changelog, store listing, privacy nutrition label / data safety form, the Accessibility Nutrition Labels, and the account-deletion requirement. Guidance-level; it never handles signing secrets. Use when asked to ship, release, submit to TestFlight / App Store / Play, or push an OTA update.

★ 0
foxinthehenhouse
foxinthehenhouse/app-in-a-box

reflect

Consolidate what recent sessions learned into the project's long-term memory (.agents/memory). Use at the end of a substantial session, after a correction from the user, after a painful debugging session, when the session-start healthcheck says reflect is overdue or a pending-reflection seed is waiting, or when MEMORY.md exceeds its budget.

★ 0
foxinthehenhouse
foxinthehenhouse/app-in-a-box

pr-review

Review a pull request on the owner's behalf and fix what's safe to fix. Runs the deterministic gates first, then correctness/security, design-system, accessibility and analytics review passes, verifies each blocking finding, applies safe fixes, re-checks, and ends with a plain-English verdict. Use when a PR is opened, when asked to review a PR, a branch or a pasted diff, or before merging anything.

★ 0
foxinthehenhouse
foxinthehenhouse/app-in-a-box

north-star-report

Weekly product-analytics and reliability readout. Pulls the funnel of the brief's 5 north-star events from PostHog, finds the biggest drop-offs and errors, checks each SLO in docs/slo.yaml against its error budget, compares with last week, and proposes (then, with a yes, files) tickets that target them. Use weekly, when the session-start line says it's overdue, when the backlog is empty, or when asked "how is the app doing?".

★ 0
foxinthehenhouse
foxinthehenhouse/app-in-a-box

next

Decide the single next best thing to work on, with two alternates. Reads setup progress, work in progress, open PRs and failing CI, the backlog (GitHub Issues or Linear), overdue harness rituals, product decisions parked until now, and north-star analytics if PostHog is connected. Use when someone asks "what next?", "what should I work on?", at the start of a session, or after a PR merges.

★ 0
foxinthehenhouse
foxinthehenhouse/app-in-a-box

new-worktree

Create an isolated git worktree + branch off origin/main for one unit of work, so parallel agents or sessions never share a checkout. Use when starting any ticket, or when another session may be working in the main checkout.

★ 0