Commit Graph

4 Commits

Author SHA1 Message Date
Clayton Kim 4115424933 Integrate worktree-agent-a3f0cd923d0e7c49f
# Conflicts:
#	evals/adversarial-evals.json
#	scripts/wiki_sync.py
2026-07-07 07:00:20 -07:00
Clayton Kim 6a8cb457d4 Cover wiki_sync's parser with an offline fixture and --from-file seam
wiki_sync.py fetched and parsed external wikitext with no eval reaching
its parser or SECTION_MAP; a drift there would silently emit wrong
diffs/prompts. Add a thin --from-file seam to check/diff/prompt that
reads wikitext from disk, skips the network fetch, and skips the sync
state write, then add an invented offline fixture and WIKI-01..03 rows
exercising the mapped-section and unmapped-section paths through it.

The fixture pins the parser, not the world; live-page drift detection
remains cmd_check. Never copy real wikitext into fixtures.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01K6CYksdLbXbTAxcAQjvHz5
2026-07-07 06:30:10 -07:00
Clayton Kim 91a99c4c9d Make the advertised Python 3.8+ floor true, and test it in CI
Insert `from __future__ import annotations` in the 17 files that use
PEP 604/585 annotations in module-level positions evaluated at import
time, so scripts/banned_phrase_scan.py and friends no longer raise
TypeError on Python 3.8/3.9. Add a 3.8 leg to the CI matrix so the
floor claim in README.md is actually gated, and correct the two
imprecise "439 deterministic cases" references to "440 deterministic
script cases (439 pass, 1 documented xfail)".

New scripts must carry the future-import until the floor is raised;
if the maintainer later chooses 3.10+, delete the CI 3.8 leg and
README claim together.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01K6CYksdLbXbTAxcAQjvHz5
2026-07-07 06:13:32 -07:00
Clayton Kim ad7fcd584f Integrate humanizer patterns + Wikipedia auto-update
Add 13 new AI-tell categories from Wikipedia's Signs of AI writing guide:
significance inflation, promotional language, superficial -ing analyses,
vague attributions, copula avoidance, formulaic challenges, negative
parallelisms, communication artifacts, knowledge-cutoff disclaimers,
generic conclusions, false ranges, elegant variation, AI vocabulary.

- taboo-phrases.md: ~150 phrases across 22 categories (up from ~87/9)
- banned_phrase_scan.py: 185 scanner entries with severity field (hard/soft)
- edit-library.md: 9 new before/after patterns (#16-24)
- personality-guide.md: new file for voice/personality guidance
- rubric.md: expanded Human Authenticity criterion
- wiki_sync.py: Wikipedia source page sync script (check/diff/prompt)
- SKILL.md: /unslop --wiki-sync maintenance command

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-08 12:59:54 -08:00