Skip to main content
hc
Submit
skillsSource-backedReview first Safety ✓ Privacy ✓

Claude Code Routines Automation Capability Pack Skill

Expert capability pack for authoring and reviewing Claude Code cloud routines via documented /schedule CLI commands, self-contained prompts, connector scoping, and post-run verification on claude.ai/code/routines.

by kiannidev·added 2026-06-16·
Claude CodeCodexWindsurfGeminiCursorCLI
HarnessClaude CodeCodexWindsurfGeminiCursorCLI
Level:expertType:capability-packVerified:validated
Review first — review before installing

Open the source and read safety notes before installing.

Safety notes

  • Routines run as full autonomous cloud sessions with no approval prompts during a run.
  • Included MCP connectors grant write tools without interactive approval—remove unused connectors.
  • Unrestricted branch pushes allow modifying non-claude/ branches—enable only when explicitly approved.

Privacy notes

  • Routine prompts and run transcripts may include proprietary code and connector payloads.
  • API trigger bearer tokens are secrets shown once on the web—store in a secret manager.
  • Actions through GitHub or connectors appear as your linked identity.

Prerequisites

  • Claude Pro, Max, Team, or Enterprise with Claude Code on the web and routines enabled.
  • claude.ai subscription login in CLI (not Console API key-only auth).
  • GitHub repositories authorized for cloud routine sessions.
  • Human reviewer for autonomous routine runs before merging branches.

Schema details

Install type
package
Reading time
8 min
Difficulty score
68
Troubleshooting
Yes
Breaking changes
No
Source repository stats
Scope
Source repo
Package metadata
Skill and platform metadata
Skill type
capability-pack
Skill level
expert
Verification
validated
Verified at
2026-06-16
Retrieval sources
https://code.claude.com/docs/en/routineshttps://code.claude.com/docs/en/skillshttps://code.claude.com/docs/en/features-overviewhttps://github.com/anthropics/claude-codehttps://developers.google.com/search/docs/fundamentals/creating-helpful-content
Tested platforms
Claude CodeClaudeCursorGeneric AGENTS
PlatformSupportInstall path
claude-codeNative.claude/skills/<skill-name>/SKILL.md
codexNative.agents/skills/<skill-name>/SKILL.md
windsurfNative.windsurf/skills/<skill-name>/SKILL.md
geminiNative.gemini/skills/<skill-name>/SKILL.md or .agents/skills/<skill-name>/SKILL.md
cursorAdapter.cursor/rules/<skill-name>.mdc
cliManualAGENTS.md or tool-specific context file
Full copyable content
# Trigger
"Apply the Claude Code routines automation capability pack for this routine."

# Required output
1) Self-contained routine prompt review
2) Repository and connector scope checklist
3) /schedule CLI management plan
4) Post-run verification steps on claude.ai/code/routines
5) Privacy-safe routine summary

About this resource

Knowledge Freshness

Grounded in Claude Code routines documentation verified on 2026-06-16. Routines are in research preview—API surfaces, limits, and CLI behavior may change; confirm live docs before production automation.

Retrieval Sources

Source Verification Notes

Verified against official routines documentation on 2026-06-16:

  • Routines are saved configurations (prompt, repositories, environment, triggers) that run on Anthropic-managed cloud infrastructure.
  • /schedule in the CLI creates scheduled routines conversationally; /schedule list, /schedule update, and /schedule run manage them.
  • /schedule requires a claude.ai subscription login—Console API keys and cloud provider auth hide the command per troubleshooting docs.
  • Routines run without approval prompts; scope repositories, network, and connectors narrowly.
  • A green run status means no infrastructure error—not guaranteed task success; reviewers must open the run transcript on claude.ai/code/routines.
  • API and GitHub triggers are configured on the web; CLI /schedule creates schedule triggers only.

Scope Note

Community capability pack for CLI schedule authoring and run review—not an official Anthropic product. Web-only trigger setup remains on claude.ai/code/routines.

Core Workflow

  1. Confirm /schedule is available (claude.ai login, CLI version, org policy).
  2. Draft a self-contained prompt stating inputs, steps, and success criteria.
  3. Select only repositories the routine must touch; default to claude/ branch pushes.
  4. Remove unused MCP connectors from the routine form defaults.
  5. Create the routine with /schedule or the web form; record the routine name.
  6. Manage cadence with /schedule list and /schedule update when needed.
  7. Trigger validation with /schedule run or Run now on the web.
  8. Open the run session and verify task outcome—not only the green status indicator.
  9. Publish a privacy-safe summary for stakeholders.

Capability Scope

  • Self-contained routine prompt review.
  • Repository and connector least-privilege scoping.
  • Documented /schedule CLI management commands.
  • Post-run verification on claude.ai/code/routines.
  • Troubleshooting /schedule Unknown command cases from official docs.

Compatibility

Native

  • Claude Code: use when authoring cloud routines from the CLI.

Manual Adaptation

  • Generic AGENTS: apply prompt and scope checklist when designing scheduled cloud agents.

Required Inputs

  • Maintenance or automation task description.
  • Target GitHub repositories and branch push policy.
  • Connector list required for the task.
  • Review owner before merging routine-opened pull requests.

Production Rules

  • Never store API trigger bearer tokens in repository files or public CI logs.
  • Trim connectors to the minimum tools required for the task.
  • Treat routine output as actions under your linked GitHub and connector identities.
  • Pause schedules with the web toggle before changing prompts in production repos.
  • Reference official docs for API /fire details—configure tokens on the web only.

Review Matrix

Check Pass criteria Doc basis
Prompt self-contained No live chat context required Create a routine
CLI auth valid /schedule available with claude.ai login Troubleshooting section
Connectors trimmed Unused MCP connectors removed Connectors section
Branch policy claude/ prefix unless unrestricted pushes approved Branch permissions
Run verified Transcript reviewed after green status View and interact with runs

Output Contract

  1. Prompt review with success criteria.
  2. Repository and connector scope notes.
  3. /schedule management command plan.
  4. Post-run verification checklist.
  5. Privacy-safe routine summary.

Troubleshooting

Issue: /schedule returns Unknown command Fix: Confirm claude.ai subscription login, CLI version at least v2.1.81, and org routines policy per official troubleshooting.

Issue: Green run status but task failed Fix: Open the run transcript on claude.ai/code/routines—docs state green means no infrastructure error only.

Duplicate Check

Distinct from routines-for-recurring-claude-code-maintenance (step-by-step guide covering web triggers) and claude-code-desktop-scheduled-maintenance-capability-pack (local Desktop scheduled tasks). This pack focuses on CLI /schedule authoring, management commands, and run verification matrices.

Editorial Disclosure

Independent entry by kiannidev based on public Claude Code documentation. No paid placement or affiliate links.

#claude-code#routines#automation#schedule#capability-pack

Source citations

Add this badge to your README

Show that Claude Code Routines Automation Capability Pack Skill is listed on HeyClaude. Paste this Markdown into your README — it renders the badge and links back to this page.

Listed on HeyClaude
[![Listed on HeyClaude](https://heyclau.de/badge/skills/claude-code-routines-automation-capability-pack.svg)](https://heyclau.de/entry/skills/claude-code-routines-automation-capability-pack)

How it compares

Claude Code Routines Automation Capability Pack Skill side by side with 3 alternatives on trust, install, platform support, and disclosed safety notes — all from reviewed registry metadata.

FieldClaude Code Routines Automation Capability Pack Skill

Expert capability pack for authoring and reviewing Claude Code cloud routines via documented /schedule CLI commands, self-contained prompts, connector scoping, and post-run verification on claude.ai/code/routines.

Open dossier 		Claude Agent SDK Custom Tool Authoring Capability Pack Skill

Expert Claude Agent SDK custom tool authoring capability pack for designing typed in-process tools with createSdkMcpServer, allowedTools scoping, isError handling, and privacy-safe rollout checklists aligned to official custom-tools documentation.

Open dossier 		Claude Code Computer Use GUI QA Capability Pack Skill

Expert Claude Code computer use GUI QA capability pack that applies documented computer-use MCP enablement, per-app session approval, screenshot checkpoints, and macOS permission workflows for native app validation—not an official Anthropic QA product.

Open dossier 		MCP Client Config Audit Capability Pack Skill

Expert MCP client config audit capability pack for reviewing Claude Code MCP server entries, scope placement, tool approval settings, env var secrets, and startup context load before enabling servers in production repositories.

Open dossier
Trust
Install riskReview firstReview firstReview firstReview first
Notes Safety ✓ Privacy ✓ Safety ✓ Privacy ✓ Safety ✓ Privacy ✓ Safety ✓ Privacy ✓
Categoryskillsskillsskillsskills
Sourcesource-backedsource-backedsource-backedsource-backed
Authorkiannidevkiannidevkiannidevkiannidev
Added2026-06-162026-06-162026-06-162026-06-14
Platforms
Claude CodeCodexWindsurfGeminiCursorCLI
Claude CodeCodexWindsurfGeminiCursorCLI
Claude CodeCodexWindsurfGeminiCursorCLI
Claude CodeCodexWindsurfGeminiCursorCLI
Source repo————
Safety notes✓Routines run as full autonomous cloud sessions with no approval prompts during a run. Included MCP connectors grant write tools without interactive approval—remove unused connectors. Unrestricted branch pushes allow modifying non-claude/ branches—enable only when explicitly approved.✓Tool descriptions and annotations are hints, not enforcement—validate destructive inputs in handlers. Wildcard mcp__server__* allowlists expand blast radius; prefer per-tool grants in production. Throwing from a handler stops the query; return isError for recoverable failures per custom-tools docs. In-process tools run inside your application whenever Claude calls them; gate writes explicitly.✓Computer use controls real desktop apps with per-app session approval—not sandboxed Bash. Review sentinel warnings before approving Terminal, Finder, or System Settings access. Press Esc or Ctrl+C to stop computer use and release the machine-wide lock immediately. Only one Claude Code session can hold the computer-use lock at a time.✓MCP config files can contain API keys, OAuth client secrets, and bearer tokens that must never be committed to git or pasted into public issues. Enabling write or execute MCP tools in shared repositories expands the blast radius for every contributor using Claude Code in that repo. Project-scoped servers affect all collaborators; user-scoped servers follow individual accounts across repositories unless restricted. Removing an MCP server from config does not automatically revoke OAuth tokens issued to remote vendors. This skill recommends config changes; it must not edit MCP settings or rotate secrets without explicit user approval.
Privacy notes✓Routine prompts and run transcripts may include proprietary code and connector payloads. API trigger bearer tokens are secrets shown once on the web—store in a secret manager. Actions through GitHub or connectors appear as your linked identity.✓Tool schemas, inputs, and results enter model context each turn—never embed secrets in descriptions. Large tool sets increase context usage; defer rarely used tools via tool search when scaling. Structured outputs and logs may reach host telemetry—redact customer identifiers at handler boundaries. External APIs invoked by handlers may log request metadata governed by vendor retention policies.✓Screenshots are downscaled before model upload but may still capture sensitive UI content. Your terminal window is excluded from screenshots per official documentation. Redact customer or proprietary UI details before sharing screenshot evidence externally. Other visible apps are hidden while Claude works; still minimize sensitive windows before starting.✓MCP config audits often expose internal service URLs, database hostnames, staging environment names, and account identifiers. Tool schema listings loaded at startup can reveal internal API surface area if copied into public support threads. Shared `.mcp.json` files checked into repositories may leak secrets through git history even after redaction in the current commit. Public audit summaries should describe scope and risk categories, not live credentials or full server manifests.
Prerequisites
  • Claude Pro, Max, Team, or Enterprise with Claude Code on the web and routines enabled.
  • claude.ai subscription login in CLI (not Console API key-only auth).
  • GitHub repositories authorized for cloud routine sessions.
  • Human reviewer for autonomous routine runs before merging branches.
  • Claude Agent SDK installed for TypeScript or Python with a working query loop.
  • Zod (TypeScript) or dict/JSON Schema (Python) for tool input contracts.
  • Written list of side effects each custom tool may perform and who approves them.
  • Staging environment to validate allowedTools, permissions, and error paths.
  • Interactive Claude Code session on macOS v2.1.85+ with Pro or Max plan per computer-use docs.
  • computer-use MCP server enabled through /mcp for the project.
  • macOS Accessibility and Screen Recording permissions granted when prompted.
  • Disposable QA target app or simulator without live customer data on screen.
  • Redacted access to project `.mcp.json`, user MCP settings, or enterprise MCP policy for the workspace under review.
  • A list of MCP servers currently configured and the repositories or teams that depend on them.
  • Permission to inspect Claude Code startup behavior with `/context` or equivalent MCP tool listings.
  • Platform or security stakeholder available to approve config changes affecting production repositories.
Install————
Config————
Citations
ClaimUnclaimedUnclaimedUnclaimedUnclaimed

Signals

Loading live community signals…

More like this, weekly

A short, calm digest of reviewed Claude resources. Unsubscribe any time.