Skip to content

docs: correct subagent nesting default (reversed in 2.1.219) and stale Algorithm LATEST pointer - #1655

Open
elhoim wants to merge 2 commits into
danielmiessler:mainfrom
elhoim:docs/correct-stale-facts
Open

docs: correct subagent nesting default (reversed in 2.1.219) and stale Algorithm LATEST pointer#1655
elhoim wants to merge 2 commits into
danielmiessler:mainfrom
elhoim:docs/correct-stale-facts

Conversation

@elhoim

@elhoim elhoim commented Jul 26, 2026

Copy link
Copy Markdown

Two stale-fact corrections in DOCUMENTATION/. No behaviour change.

Subagent nesting default

The default has moved twice in three releases — 5 until 2.1.217, 1 in 2.1.217, 3 from 2.1.219 — and the guidance still said subagents cannot spawn subagents, steering away from a pattern the platform now enables by default. Records the current depth-3 default with CLAUDE_CODE_MAX_SUBAGENT_SPAWN_DEPTH=1 to flatten it, plus the version history so the next move is easy to spot.

Also adds the two per-session ceilings (200 subagent spawns, 200 WebSearch calls) that a long fan-out session can actually reach.

Per your note on #1569, the concurrency-cap wording stays neutral on what happens to over-cap spawns — the changelog states the cap but not whether they queue or are refused.

Algorithm version pointer

The row that teaches how to resolve the Algorithm version named 8.1.0 while ALGORITHM/LATEST reads 8.4.0. Line 26 of the same file was already correct, so this was a single missed edit rather than systemic lag.


Two notes so you can close this fast if it is already handled:

elhoim added 2 commits July 26, 2026 17:30
The nesting default has moved twice in three releases: 5 until 2.1.217,
1 in 2.1.217, and 3 from 2.1.219. Guidance stating that subagents cannot
spawn subagents steers away from a pattern the platform now enables by
default.

Record the current depth-3 default with the flag to flatten it, and add
the two per-session ceilings (200 subagent spawns, 200 WebSearch calls)
that a long fan-out session can actually reach.

Wording on the concurrency cap stays neutral on what happens to over-cap
spawns, per the note on danielmiessler#1569 that the changelog does not state whether
they queue or are refused.
The row that teaches how to resolve the Algorithm version named 8.1.0
while ALGORITHM/LATEST reads 8.4.0 — three versions of drift in the one
place a reader looks up the answer. Line 26 of the same file was already
correct, so this was a single missed edit rather than a systemic lag.

(cherry picked from commit deac16f)
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant