docs: record 2026-08-23 audit sitting outcomes

Add a Stability Exceptions section to docs/audit-logs.md documenting
the zensical override (admitted pre-1.0 to displace a dying mkdocs
upstream). Add a matching evidence bullet to CLAUDE.md and AGENTS.md
warning that a download count is not automatically independent demand
when one listed entry depends on another, using mkdocs / mkdocs-material
as the worked example. Both facts were discovered during this sitting
and need to persist so the next audit doesn't rediscover them.

Co-Authored-By: Claude <noreply@anthropic.com>

AGENTS.md

@@ -9,6 +9,7 @@ An opinionated guide to the best Python frameworks, libraries, and tools.
 [CONTRIBUTING.md](CONTRIBUTING.md) holds the admission rules, quality requirements, rejection rules, entry format, and ordering. Apply it whenever adding or removing an entry — direct commits included, not only PR reviews.
 
 - Every keep/drop reason must be verified against current online data at decision time — download counts, repo activity and archived status, PyPI metadata, project docs. Judging tiers — obvious choice vs challenger — also requires WebSearch evidence (adoption trajectory, community sentiment), not download counts alone. Training-data recollections are not evidence; label anything unverifiable as a judgment call.
+- A download count is not automatically independent demand. When one listed entry depends on another, check `requires_dist` on PyPI before citing the depended-on entry's count: mkdocs-material hard-depends on mkdocs, so mkdocs' figure exceeded mkdocs-material's by only about one percent and nearly all of it was mkdocs-material pulling it in.
 - One entry per commit when adding or deleting entries. Exceptions: a prune sweep is one commit per section, its body listing each removal with its reason; format, wording, or categorization changes may be bundled. Cross-section re-homes ride the originating audit's commit (both sides of the move in one diff).
 - Resources sections are not project entries: out of audit scope, and the website never parses them.
 - Sponsor placement never influences which projects get listed — see [SPONSORSHIP.md](SPONSORSHIP.md).

CLAUDE.md

@@ -9,6 +9,7 @@ An opinionated guide to the best Python frameworks, libraries, and tools.
 [CONTRIBUTING.md](CONTRIBUTING.md) holds the admission rules, quality requirements, rejection rules, entry format, and ordering. Apply it whenever adding or removing an entry — direct commits included, not only PR reviews.
 
 - Every keep/drop reason must be verified against current online data at decision time — download counts, repo activity and archived status, PyPI metadata, project docs. Judging tiers — obvious choice vs challenger — also requires WebSearch evidence (adoption trajectory, community sentiment), not download counts alone. Training-data recollections are not evidence; label anything unverifiable as a judgment call.
+- A download count is not automatically independent demand. When one listed entry depends on another, check `requires_dist` on PyPI before citing the depended-on entry's count: mkdocs-material hard-depends on mkdocs, so mkdocs' figure exceeded mkdocs-material's by only about one percent and nearly all of it was mkdocs-material pulling it in.
 - One entry per commit when adding or deleting entries. Exceptions: a prune sweep is one commit per section, its body listing each removal with its reason; format, wording, or categorization changes may be bundled. Cross-section re-homes ride the originating audit's commit (both sides of the move in one diff).
 - Resources sections are not project entries: out of audit scope, and the website never parses them.
 - Sponsor placement never influences which projects get listed — see [SPONSORSHIP.md](SPONSORSHIP.md).

docs/audit-logs.md

@@ -22,6 +22,12 @@ Display names follow the canonical PyPI package name. These entries keep a diffe
 - strawberry -- `strawberry-graphql`.
 - strawberry-django -- `strawberry-graphql-django`.
 
+### Stability Exceptions
+
+Entries admitted despite the Stable quality requirement, by maintainer decision:
+
+- zensical -- admitted 2026-08-23 at version 0.0.57, PyPI classifier `Development Status :: 3 - Alpha`, with no 1.0 target announced. Admitted as a challenger displacing mkdocs because the Documentation section was carrying a dying upstream: the Material for MkDocs team's announcement of 2025-11-05 put Material into maintenance mode for twelve months and called MkDocs itself unmaintained since 2024-08 and a supply chain risk. Re-check when 1.0 lands.
+
 ### Mature-stable Keeps
 
 These entries sit past the 12-month activity requirement without an override. Each one is kept by editorial judgment: mature, stable, and no successor exists.