Skip to content

docs: correct four figures and one claim both guides had outgrown - #29

Merged
cmj0121 merged 1 commit into
mainfrom
docs/refresh-drifted-figures
Sep 16, 2026
Merged

cmj0121 merged 1 commit into
mainfrom
docs/refresh-drifted-figures

Conversation

@cmj0121

@cmj0121 cmj0121 commented Sep 16, 2026

Copy link
Copy Markdown
Owner

A sweep of every mechanically checkable number in CLAUDE.md and README.md against the repo.
Five claims were wrong, all in the same direction — written true, then overtaken by work that
had no reason to revisit them.

claim was is
description average (CLAUDE.md) 173 204
description target (README) "kept near 160 characters" 204 average, ceiling 320
level-1 cost (README) "under 800 tokens" ~1,000 — name + description is 4,063 chars
allowed-tools grants with a space 70 79 of 173
references/ machinery (README) "has so far gone unused" used by references/iteration.md

The first three share one cause: #14 and #22 gave every description a negative boundary clause.
That was the trade that bought fewer misfires, so the paragraphs now name it rather than quoting
a figure from before it. The grant count is now stated as a share of the total, so the next
drift is visible inside the sentence.

The fifth is the one that was not just a number. Both guides used Phase 2.5 as the example
of what belongs in references/, and README still said the machinery had gone unused. Neither
holds: what actually moved is Phase 5, into references/iteration.md, because Autonomous
mode is the only mode that runs it. Phase 2.5 is now named as the counter-example it really is
— conditional, but eight lines, most of them the condition itself, so a fetch would cost more
than it saves. That matters because the two paragraphs are the repo's rule for what moves;
illustrating it with a case that fails its own second test taught the wrong thing.

Verified and left alone

19 plugins and skills · 380 queries at 20 per plugin, 10/10 labels, 12/8 train/validation ·
13 eval cases at runs: 3 · five verdict markers · the 320 / 500 / 300 / 100 thresholds ·
the model-tier roster (7 pins, 4 with context: fork + background: false) ·
spec-writer's 5000 words / 800 lines.

No plugin changed, so nothing bumps: neither behaviour nor catalogue copy.
make test 4/4, scripts/validate-fixtures and make doctor clean.

🤖 Generated with Claude Code

A sweep of every mechanically checkable number in CLAUDE.md and README.md
against the repo. Five were wrong, all in the same direction -- written true,
then overtaken by work that had no reason to revisit them.

- description average 173 -> 204, and README's "kept near 160 characters" with
  it. #14 and #22 gave every description a negative boundary clause; that is
  the trade that bought fewer misfires, so the paragraph now names it rather
  than quoting a figure from before it.
- README's "all 19 together cost a session under 800 tokens" -> about 1,000.
  Same cause: name + description is 4,063 characters.
- allowed-tools grants containing a space, 70 -> 79, now stated as a share of
  the 173 total so the next drift is visible in the sentence itself.
- The references/ example was Phase 2.5 in both guides, and README still said
  the machinery had "gone unused". Neither is true: what moved is Phase 5, into
  references/iteration.md, because Autonomous mode is the only mode that runs
  it. Phase 2.5 is named as the counter-example it actually is -- conditional,
  but eight lines, so a fetch costs more than it saves.

Verified and left alone: 19 plugins and skills, 380 queries at 20 per plugin
with a 10/10 label split and a 12/8 train/validation split, 13 eval cases at
runs: 3, five verdict markers, the 320/500/300/100 thresholds, the model-tier
roster, and spec-writer's 5000 words / 800 lines.

No plugin changed, so nothing bumps -- neither behaviour nor catalogue copy.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
@cmj0121
cmj0121 merged commit ce0aa0f into main Sep 16, 2026
2 checks passed
@cmj0121
cmj0121 deleted the docs/refresh-drifted-figures branch September 16, 2026 22:27
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