docs(CLAUDE.md): discourage unprompted subagents and dynamic workflows#24027
Merged
Conversation
AztecBot
added
ci-draft
ludamad
approved these changes
Jun 11, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
v5-next counterpart of #24026.
Adds an
<agent_and_workflow_restraint>block to the rootCLAUDE.mdtelling Claude to do work inline in the current session and not spawn parallel subagents or launch dynamic workflows unless the user explicitly asks.Why
Operators have reported burning through their token budget from a single prompt that quietly fanned out — in one case a "summarize recent ZK advancements" query started ~30 agents, and another exhausted a 5h budget spinning up subagents. Parallel agents and dynamic workflows multiply spend (≈2x for one helper, far more for a swarm) and the user can neither see the fan-out coming nor stop it. This appears to be a current tendency of Fable. The guidance reasserts: handle search/summarize/research/multi-file edits inline, reserve subagents for explicit user requests or a single read-heavy isolation case, and never start a dynamic workflow by default.
Passes the repo's
<editorial_test>: the line would have prevented the ~30-agent fan-out on an ordinary research prompt described above.An equivalent shared rule is being added in the claudebox repo so it applies to every managed session.
Created by claudebox · group:
slackbot