claude-code - 💡(How to fix) Fix Allow disabling built-in skills (e.g. `schedule`) via settings [1 comments, 2 participants]

Official PRs (…)
ON THIS PAGE

Recommended Tools

×6

Utilities matched from this issue’s tags and category — try them while you read without losing context.

GitHub issue graph ai analysis

Paste a GitHub issue URL. We fetch that issue, discover linked issues from bodies/comments/timeline, collect linked pull requests, and produce a structured English report.

The report is written in English Markdown for sharing and archival.

Helpful · Quick feedback

Loading…
GitHub stats
anthropics/claude-code#53746Fetched 2026-04-28 06:48:15
View on GitHub
Comments
1
Participants
2
Timeline
5
Reactions
0
Timeline (top)
labeled ×3commented ×1cross-referenced ×1

Code Example

{
  "disabledBuiltinSkills": ["schedule"]
}
RAW_BUFFERClick to expand / collapse

Problem

Built-in skills are injected into the system prompt of every Claude Code session. The schedule skill (which exposes ScheduleWakeup, CronCreate, CronList, CronDelete, plus the /schedule slash-command framing) costs roughly 400 tokens per session.

For users or organizations that have cloud agents disabled via managed settings, /schedule and the associated tools cannot be invoked, but the skill text is still present in the system prompt. There is currently no way to opt out of it.

Proposed solution

A settings key — honored in both user settings.json and managed-settings.json — to disable named built-in skills:

{
  "disabledBuiltinSkills": ["schedule"]
}

Behavior:

  • Disabled skills' system-prompt sections are omitted.
  • Their slash commands (/schedule) and tools (ScheduleWakeup, CronCreate, …) are unregistered, so the model can't try to call them.

Relationship to existing issues

  • #52802 proposes overriding named directive sections of the core system prompt (tone, output formatting, plan-mode framing). Its author explicitly scopes out skill injection, so it does not cover this case.
  • PR #46024 (--exclude-dynamic-system-prompt-sections) is print-mode only and does not target skills.

extent analysis

TL;DR

To address the issue of the schedule skill being injected into the system prompt and consuming tokens, even when cloud agents are disabled, consider adding a settings key to disable named built-in skills.

Guidance

  • Introduce a disabledBuiltinSkills key in both settings.json and managed-settings.json to allow users and organizations to opt out of specific skills.
  • Set the value of disabledBuiltinSkills to an array containing the names of skills to be disabled, such as ["schedule"].
  • Ensure that when a skill is disabled, its system-prompt sections are omitted and its associated slash commands and tools are unregistered.
  • Review related issues, such as #52802 and PR #46024, to understand the broader context and potential overlaps with the proposed solution.

Example

{
  "disabledBuiltinSkills": ["schedule"]
}

Notes

The proposed solution focuses on adding a settings key to disable specific built-in skills. This approach may not address all aspects of skill injection and token consumption but provides a targeted way to opt out of the schedule skill.

Recommendation

Apply the workaround by introducing the disabledBuiltinSkills key in the settings files, as it directly addresses the issue of unnecessary token consumption by the schedule skill when cloud agents are disabled.

Vote matrix · Quick signals

Works
Did the solution work? Tap to confirm.
Easy Fix
Was it a quick fix?
Time Saver
Did it save you time?
Blocking
Was it severely blocking?
Common Issue
Are others likely hitting this too?
Flaky / Intermittent
Is it intermittent?
Verified / Reproducible
Can you reproduce it reliably?
Loading…

Still need to ship something?

×6

Another batch ranked right after the header list — different links, same matching logic.

Back to top recommendations

TRENDING