Skip to content

Kartik-Doye/Ravengard-AI-Recruiter / .claude/skills/not-ai

not-ai

Verified

Edit prose into a clear, specific, source-grounded version that preserves the author's meaning and voice. Use for humanizing stiff writing, removing generic phrasing, matching a supplied voice sample, diagnosing robotic prose, or drafting from the user's notes. Do not optimize for AI-detector scores or claim to prove authorship.

FileKindSize
SKILL.mdskill md14.1 KB
reference/mechanical-tells.mdmarkdown1.5 KB
reference/profile.mdmarkdown1.7 KB
reference/research-sources.mdmarkdown2.1 KB
reference/vocabulary.mdmarkdown1.6 KB
reference/why-word-swapping-fails.mdmarkdown1.4 KB

reference/research-sources.md

View on GitHub
# Research sources and limits

Read this reference when explaining why the skill reviews genre fit,
grammatical density, engagement, stock wording, or detector claims.

## Main sources

- Reinhart, Alex, et al. [*Do LLMs write like humans? Variation in grammatical
  and rhetorical styles*](https://doi.org/10.1073/pnas.2422455122). *PNAS*,
  2025. Parallel human and model continuations across several English genres,
  analyzed with 66 Biber features. The authors present the differences as
  revision and teaching opportunities, not an individual authorship test.
- Jiang, Feng Kevin, and Ken Hyland. [*Does ChatGPT write like a student?
  Engagement markers in argumentative essays*](https://doi.org/10.1177/07410883251328311).
  *Written Communication*, 2025. In that task and corpus, student essays used a
  richer range of reader-engagement features.
- Kobak, Dmitry, et al. [*Delving into LLM-assisted writing in biomedical
  publications through excess vocabulary*](https://doi.org/10.1126/sciadv.adt3813).
  *Science Advances*, 2025. A corpus-level study of vocabulary shifts in
  biomedical abstracts; not a verdict on individual documents.
- [Wikipedia: Signs of AI Writing](https://en.wikipedia.org/wiki/Wikipedia:Signs_of_AI_writing).
  A changing, context-specific field guide and not a Wikipedia policy. The page
  itself warns that its signs are descriptive and not proof.
- Sadasivan, Vinu Sankar, et al. [*Can AI-Generated Text be Reliably
  Detected?*](https://arxiv.org/abs/2303.11156), 2023. Shows practical and
  theoretical limits of text-only detection under distribution overlap and
  paraphrasing.
- Liang, Weixin, et al. [*GPT detectors are biased against non-native English
  writers*](https://doi.org/10.1016/j.patter.2023.100779). *Patterns*, 2023.
  Documents serious false positives for the evaluated non-native English
  samples and detectors.

## Product interpretation

These sources justify contextual editorial review. They do not justify fixed
feature targets, word bans, claims of authorship, or detector optimization.
Findings depend on model, prompt, language, register, corpus, and time. Keep the
writer's source and reader's task above every proxy.
slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack."Stars: 198kUpdated Claude CodeCodexVerifiedInstall
ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read.Stars: 194kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report.Stars: 194kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt.Stars: 194kUpdated Claude CodeCodexVerifiedInstall