# AI use cases for Humanizer Stack (NulightJens/humanizer-stack) Repo: [NulightJens/humanizer-stack](https://github.com/NulightJens/humanizer-stack) | Use cases: 4 | Page: https://dedupedai.com/use-cases/nulightjens-humanizer-stack/ A two-pass set of Claude Code Skills and Python scanners that removes AI writing tells from outward-facing text, first at the word level and then at the structural level. This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL. - Strip AI tells from a blog post before publishing: https://dedupedai.com/use-cases/nulightjens-humanizer-stack/strip-ai-tells-before-publishing-post/ - Block AI-tell copy from merging with a strict scan: https://dedupedai.com/use-cases/nulightjens-humanizer-stack/gate-drafts-with-strict-copy-scan/ - Run the structural pass on a long draft or story: https://dedupedai.com/use-cases/nulightjens-humanizer-stack/run-structural-pass-on-long-draft/ - Find a shared skeleton across a series of drafts: https://dedupedai.com/use-cases/nulightjens-humanizer-stack/check-series-for-shared-skeleton/ ## How to strip AI tells from a blog post before publishing with Humanizer Stack Repo: [NulightJens/humanizer-stack](https://github.com/NulightJens/humanizer-stack) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Run the word-level pass and scanner on an AI-drafted post so it reads like a person wrote it. **Who it is for:** A founder or content marketer who has an AI-drafted post that needs to sound like their own voice before it goes live. **The problem:** The draft carries word-level tells such as em dash overuse, hype vocabulary, and negative parallelism that readers notice. **The outcome:** A revised draft with the flagged tells fixed, plus a short list of scanner hits the author reviewed and kept on purpose. ### Steps 1. `git clone https://github.com/NulightJens/humanizer-stack.git` 2. `cd humanizer-stack` 3. `./install.sh` 4. `python3 scripts/copy_scan.py draft.md` 5. In Claude Code, send: `humanize this post`, then rerun `python3 scripts/copy_scan.py draft.md` to confirm the hits are gone or marked as intentional. ### Prompt to paste into your agent ```text Using the humanizer-stack repo at , run python3 scripts/copy_scan.py on and then apply the humanizer skill to fix every flagged word-level tell. Output the revised file and any remaining scanner hits. Verify that the rerun of the scanner shows no hits except lines the author marked copy-ignore. ``` ### Caveat The scanners catch only the pattern-matchable slice, so a human reader still has to judge cadence and polished but empty filler. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/nulightjens-humanizer-stack/strip-ai-tells-before-publishing-post/ ## How to block AI-tell copy from merging with a strict scan with Humanizer Stack Repo: [NulightJens/humanizer-stack](https://github.com/NulightJens/humanizer-stack) | Difficulty: Advanced | Time: 30 minutes | Basis: From the README Use the scanner's exit codes and JSON output to fail a check when a draft still has copy tells. **Who it is for:** A developer or operations lead who keeps published copy in a repo and wants a repeatable check before changes are merged. **The problem:** Word-level tells slip into shared copy files during edits, and nobody runs the check by hand every time. **The outcome:** A scan step that returns a nonzero exit code on any hit and produces machine-readable output for review tooling. ### Steps 1. `git clone https://github.com/NulightJens/humanizer-stack.git` 2. `cd humanizer-stack` 3. `python3 scripts/copy_scan.py --json draft.md` 4. `python3 scripts/copy_scan.py --strict draft.md` 5. Add the `--strict` command to a pre-commit or CI step that runs on the copy files, and mark intentional usages with `copy-ignore`. (inferred, check the README) ### Prompt to paste into your agent ```text In the humanizer-stack repo at , run python3 scripts/copy_scan.py --strict on and capture the exit code and the output from --json. Report each hit with its line number and whether the file passed. Verify that a file with no hits returns exit code 0. ``` ### Caveat The scanner is hook-friendly but only catches the pattern-matchable slice, so it cannot replace the judgment work in the skills. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/nulightjens-humanizer-stack/gate-drafts-with-strict-copy-scan/ ## How to run the structural pass on a long draft or story with Humanizer Stack Repo: [NulightJens/humanizer-stack](https://github.com/NulightJens/humanizer-stack) | Difficulty: Intermediate | Time: 45 minutes | Basis: From the README Check discourse structure with six audits, then vary the shape so the piece is not a new template. **Who it is for:** A fiction writer or long-form author whose draft reads as if it explains its theme, resolves every thread, and names every feeling. **The problem:** Surface edits fix the words but leave the skeleton of the piece intact, and the structure is where the durable fingerprint lives. **The outcome:** A structural scan report and a revised draft with one or two interventions chosen on purpose for that piece. ### Steps 1. `git clone https://github.com/NulightJens/humanizer-stack.git` 2. `cd humanizer-stack` 3. `./install.sh` 4. `python3 skills/structural-humanizer/scripts/structural_scan.py draft.md` 5. In Claude Code, send: `run the structural pass on draft.md`, and read `skills/structural-humanizer/references/genre-calibration.md` to confirm which audits apply to your genre. ### Prompt to paste into your agent ```text Using the humanizer-stack repo at , run the structural_scan.py script on and then apply the structural-humanizer skill audits that apply to this genre. Output a revised draft and a list of the audits that changed text. Verify that at most two interventions were applied in this piece and that each one is explained. ``` ### Caveat The study covered roughly 5,000-word fiction, so applying the audits to short nonfiction is an inference; audits 1, 3, 4, and 6 transfer most cleanly. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/nulightjens-humanizer-stack/run-structural-pass-on-long-draft/ ## How to find a shared skeleton across a series of drafts with Humanizer Stack Repo: [NulightJens/humanizer-stack](https://github.com/NulightJens/humanizer-stack) | Difficulty: Intermediate | Time: 45 minutes | Basis: From the README Scan several recent pieces so a new draft does not repeat the shape of the last three. **Who it is for:** A content lead publishing a regular series of posts or lessons who notices every piece opens and closes the same way. **The problem:** Structural convergence makes a series detectable and flat, and it is hard to see from inside a single draft. **The outcome:** A side-by-side view of the structural scan results for recent pieces, with a note on which shape each new draft should take. ### Steps 1. `git clone https://github.com/NulightJens/humanizer-stack.git` 2. `cd humanizer-stack` 3. `python3 skills/structural-humanizer/scripts/structural_scan.py post-1.md` (inferred, check the README) 4. Run the same command on the last three published pieces and compare the results, focusing on audit 6, shape convergence. 5. In Claude Code, send: `run the structural pass on draft.md`, then vary one or two interventions from the earlier pieces. ### Prompt to paste into your agent ```text Using the humanizer-stack repo at , run structural_scan.py on each file in and compare the results for audit 6, shape convergence. Output a short table of the shared patterns and one suggested shape for that differs from them. Verify that the suggestion changes at least one structural choice and does not add a new fixed template. ``` ### Caveat The scanner only flags grep-able structural tells, so judging whether a whole series shares one skeleton still needs a human reader. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/nulightjens-humanizer-stack/check-series-for-shared-skeleton/