diff --git a/docs/superpowers/contracts/archive-plagiarism-tools-2026-08-18.json b/docs/superpowers/contracts/archive-plagiarism-tools-2026-08-18.json
new file mode 100644
index 00000000..daf18ca3
--- /dev/null
+++ b/docs/superpowers/contracts/archive-plagiarism-tools-2026-08-18.json
@@ -0,0 +1,16 @@
+{
+  "title": "Move install_plagiarism_tools.sh and siblings to testsAndMisc-archive",
+  "objective": "The user said they no longer use the plagiarism-detection tooling (built and just split this session per NEXT_SESSION_PROMPT.md §3). Move it out of testsAndMisc into testsAndMisc-archive with full git history preserved, rather than leaving unused code in the primary repo.",
+  "acceptance_criteria": [
+    "All 13 plagiarism-tool files (install_plagiarism_tools.sh, download_nltk_data.py, check_plagiarism.py, plagiarism_similarity.py, sherlock.py, the 3 lib/plagiarism_*.sh siblings, and the 5 lib/tests/*plagiarism* test files) exist in kuhyx/testsAndMisc-archive with their git history intact, including the pre-rename path (linux_configuration/scripts/utils/install_plagiarism_tools.sh) from before the single_use/periodic_background reorg.",
+    "The archive push completed and is verified live on GitHub before any deletion happens in the main repo -- deleting first and discovering a failed archive push would lose the only copy.",
+    "The same 13 files are removed from testsAndMisc via git rm, with no other files touched.",
+    "check_file_length.sh --all reports the same 14 over-cap files as immediately after the split commit (2a7b9d9) -- this deletion does not change that count, since none of the removed files were over cap.",
+    "pre-commit run --files passes on the deletion commit's staged set."
+  ],
+  "out_of_scope": [
+    "Any other unused or academic-tooling files (generate_study_materials.sh, repo_to_study.sh, etc.) -- user explicitly scoped this to just the plagiarism-tool files.",
+    "Re-running or re-verifying the split work itself; that was already proven correct before the archival decision was made (see install-plagiarism-tools-split-2026-08-18.json)."
+  ],
+  "verifier": "gh api repos/kuhyx/testsAndMisc-archive/contents/linux_configuration/scripts/single_use/utils confirming all 13 files present before the git rm commit; git log --oneline -- <each removed path> in testsAndMisc-archive confirming history depth matches the pre-archival testsAndMisc log; check_file_length.sh --all; pre-commit run --files on the deletion commit."
+}
diff --git a/docs/superpowers/evidence/archive-plagiarism-tools-2026-08-18.json b/docs/superpowers/evidence/archive-plagiarism-tools-2026-08-18.json
new file mode 100644
index 00000000..7b0d84fd
--- /dev/null
+++ b/docs/superpowers/evidence/archive-plagiarism-tools-2026-08-18.json
@@ -0,0 +1,43 @@
+{
+  "intent": "Move the just-split, unused plagiarism-detection tooling (install_plagiarism_tools.sh and its 12 extracted/test siblings) out of testsAndMisc into testsAndMisc-archive with full git history, per the user's explicit statement that they no longer use it.",
+  "scope": [
+    "linux_configuration/scripts/single_use/utils/install_plagiarism_tools.sh, download_nltk_data.py, check_plagiarism.py, plagiarism_similarity.py, sherlock.py (deleted)",
+    "linux_configuration/scripts/single_use/utils/lib/plagiarism_python.sh, plagiarism_sherlock.sh, plagiarism_optional.sh (deleted)",
+    "linux_configuration/scripts/single_use/utils/lib/tests/plagiarism_harness.sh, test_plagiarism_python.sh, test_plagiarism_sherlock.sh, test_plagiarism_optional.sh, test_install_plagiarism_tools.sh (deleted)",
+    "kuhyx/testsAndMisc-archive: same 13 files added, with history",
+    "Not in scope: any other unused academic-tooling file the user did not name (e.g. generate_study_materials.sh)"
+  ],
+  "changes": [
+    "Cloned testsAndMisc into a throwaway scratchpad location (never a worktree of the real repo, since git filter-repo rewrites refs and a worktree shares the object store).",
+    "Ran git filter-repo --force with explicit --path for all 13 current-path files plus the pre-rename path (linux_configuration/scripts/utils/install_plagiarism_tools.sh, from before the single_use/periodic_background reorg) so history predating that rename was not silently dropped.",
+    "Cloned testsAndMisc-archive separately, added the filtered clone as a remote, and merged it in with --allow-unrelated-histories (required since the two repos share no common root commit).",
+    "Verified the merge landed all 13 files with git ls-tree, then pushed to testsAndMisc-archive and confirmed live via the GitHub API before touching the main repo.",
+    "Only after the archive push was confirmed live: git rm the same 13 files from testsAndMisc and committed the deletion.",
+    "Removed the plag remote from the archive-clone before pushing, so no stray remote-tracking branches from the filtered clone's other refs (copilot-wt, etc.) leaked into the push."
+  ],
+  "verification": [
+    {
+      "command": "gh api repos/kuhyx/testsAndMisc-archive/contents/linux_configuration/scripts/single_use/utils",
+      "result": "pass",
+      "evidence": "Returned check_plagiarism.py, download_nltk_data.py, install_plagiarism_tools.sh, lib, plagiarism_similarity.py, sherlock.py -- confirmed live on GitHub before the main-repo deletion commit was made."
+    },
+    {
+      "command": "git log --oneline -- linux_configuration/scripts/single_use/utils/install_plagiarism_tools.sh (in the archive clone) vs the same command in testsAndMisc at commit 2a7b9d9",
+      "result": "pass",
+      "evidence": "Both show 4 commits with identical messages in the same order (the split, the no-inline-Python gate, the shellcheck fail-build change, the single_use/periodic_background rename) -- history genuinely carried over, not truncated at the rename."
+    },
+    {
+      "command": "bash meta/scripts/check_file_length.sh --all",
+      "result": "pass",
+      "evidence": "14 files over cap, unchanged from immediately after the split commit -- none of the deleted files were over cap, so this operation does not move that number."
+    }
+  ],
+  "risks": [
+    "The archive repo has no pre-commit hooks or CI, so nothing re-verifies the moved code still works there -- it is frozen at exactly the state it left testsAndMisc in.",
+    "git filter-repo was run with --force on a disposable clone rather than a truly fresh `git clone` from the remote; verified this clone's origin was the local repo path (harmless) and that no push was made to it, but noting the deviation from the tool's documented 'fresh clone' expectation."
+  ],
+  "rollback": [
+    "In testsAndMisc: git revert the deletion commit to restore the 13 files (their content and permissions are unchanged from the split commit, so a straight revert is safe).",
+    "The testsAndMisc-archive copy can stay regardless of rollback -- having the files in both places briefly is not harmful, just redundant. If it must be undone: git revert the merge commit there, though a merge revert needs the -m flag to pick which parent to keep."
+  ]
+}
diff --git a/linux_configuration/scripts/single_use/utils/check_plagiarism.py b/linux_configuration/scripts/single_use/utils/check_plagiarism.py
deleted file mode 100755
index 71bb08d0..00000000
--- a/linux_configuration/scripts/single_use/utils/check_plagiarism.py
+++ /dev/null
@@ -1,165 +0,0 @@
-#!/usr/bin/env python3
-"""Compare documents for similarity using TF-IDF cosine similarity.
-
-Extracted from `install_plagiarism_tools.sh`, which used to write this out via
-a `cat <<'PYEOF'` heredoc. Inline heredoc Python is invisible to ruff, mypy,
-pylint, bandit and pytest, so it lives in its own module instead, matching
-`download_nltk_data.py`'s extraction. The comparison engine (file reading,
-preprocessing, TF-IDF scoring) lives in `plagiarism_similarity.py`, split out
-to keep both files under the 250-line cap.
-
-The installer copies this file into `$INSTALL_DIR/check_plagiarism.py` and
-wires the `plagcheck` wrapper to run it inside the venv.
-
-Usage:
-    python check_plagiarism.py file1.txt file2.txt [file3.txt ...]
-    python check_plagiarism.py --dir /path/to/documents/
-"""
-
-from __future__ import annotations
-
-import argparse
-from pathlib import Path
-import sys
-
-from plagiarism_similarity import (
-    compute_similarity_matrix,
-    find_similar_passages,
-    preprocess_text,
-    read_file,
-)
-
-#: Minimum number of documents needed for a pairwise comparison to make sense.
-MIN_DOCUMENTS = 2
-
-
-def _out(line: str = "") -> None:
-    sys.stdout.write(line + "\n")
-
-
-def _collect_files(args: argparse.Namespace) -> list[str]:
-    files: list[str] = list(args.files)
-    if args.dir:
-        dir_path = Path(args.dir)
-        for ext in ("*.txt", "*.pdf", "*.docx", "*.md", "*.tex"):
-            files.extend(str(f) for f in dir_path.glob(ext))
-    return files
-
-
-def _read_documents(files: list[str]) -> tuple[list[str], list[str]]:
-    documents: list[str] = []
-    filenames: list[str] = []
-    for f in files:
-        if not Path(f).exists():
-            _out(f"Warning: {f} does not exist")
-            continue
-        text = read_file(f)
-        if text.strip():
-            documents.append(preprocess_text(text))
-            filenames.append(Path(f).name)
-        else:
-            _out(f"Warning: {f} is empty or unreadable")
-    return documents, filenames
-
-
-def _report_pairs(
-    documents: list[str],
-    filenames: list[str],
-    sim_matrix: list[list[float]],
-    threshold: float,
-) -> list[tuple[int, int, float, str]]:
-    _out(f"{'Document Pair':<50} {'Similarity':>12}")
-    _out("-" * 62)
-
-    suspicious_pairs: list[tuple[int, int, float, str]] = []
-    for i in range(len(documents)):
-        for j in range(i + 1, len(documents)):
-            similarity = sim_matrix[i][j]
-            pair_name = f"{filenames[i]} <-> {filenames[j]}"
-            if similarity >= threshold:
-                suspicious_pairs.append((i, j, similarity, pair_name))
-                _out(f"{pair_name:<50} {similarity:>10.1%} ⚠️")
-            else:
-                _out(f"{pair_name:<50} {similarity:>10.1%}")
-
-    _out("-" * 62)
-    return suspicious_pairs
-
-
-def _print_detailed_passages(
-    files: list[str], suspicious_pairs: list[tuple[int, int, float, str]]
-) -> None:
-    _out("\n" + "=" * 60)
-    _out(" Detailed Similar Passages")
-    _out("=" * 60)
-
-    for i, j, sim, pair_name in suspicious_pairs[:3]:
-        _out(f"\n{pair_name} ({sim:.1%} similar):")
-        _out("-" * 40)
-
-        raw_docs = [read_file(files[i]), read_file(files[j])]
-        passages = find_similar_passages(raw_docs[0], raw_docs[1])
-
-        for s1, s2, psim in passages[:5]:
-            _out(f'\n[{psim:.0%}] Document 1: "{s1[:100]}..."')
-            _out(f'      Document 2: "{s2[:100]}..."')
-
-
-def main() -> int:
-    """Compare every pair of input documents and report suspicious pairs."""
-    parser = argparse.ArgumentParser(
-        description="Text Plagiarism Checker - Compare documents for similarity"
-    )
-    parser.add_argument("files", nargs="*", help="Files to compare")
-    parser.add_argument("--dir", "-d", help="Directory containing documents to compare")
-    parser.add_argument(
-        "--threshold",
-        "-t",
-        type=float,
-        default=0.3,
-        help="Similarity threshold for flagging (0-1, default: 0.3)",
-    )
-    parser.add_argument(
-        "--detailed", "-v", action="store_true", help="Show detailed similar passages"
-    )
-
-    args = parser.parse_args()
-
-    files = _collect_files(args)
-    if len(files) < MIN_DOCUMENTS:
-        _out("Error: Need at least 2 files to compare")
-        parser.print_help()
-        return 1
-
-    _out(f"\n{'=' * 60}")
-    _out(f" Plagiarism Check - Analyzing {len(files)} documents")
-    _out(f"{'=' * 60}\n")
-
-    documents, filenames = _read_documents(files)
-    if len(documents) < MIN_DOCUMENTS:
-        _out("Error: Not enough valid documents to compare")
-        return 1
-
-    _out("Computing document similarities...\n")
-    sim_matrix = compute_similarity_matrix(documents)
-
-    suspicious_pairs = _report_pairs(documents, filenames, sim_matrix, args.threshold)
-
-    if suspicious_pairs:
-        _out(
-            f"\n⚠️  {len(suspicious_pairs)} pair(s) exceed "
-            f"{args.threshold:.0%} similarity threshold\n"
-        )
-        if args.detailed:
-            _print_detailed_passages(files, suspicious_pairs)
-    else:
-        _out(f"\n✓ No document pairs exceed {args.threshold:.0%} similarity threshold")
-
-    _out("\n" + "=" * 60)
-    _out(" Analysis complete")
-    _out("=" * 60 + "\n")
-    return 0
-
-
-if __name__ == "__main__":
-    raise SystemExit(main())
diff --git a/linux_configuration/scripts/single_use/utils/download_nltk_data.py b/linux_configuration/scripts/single_use/utils/download_nltk_data.py
deleted file mode 100755
index 0d3064a5..00000000
--- a/linux_configuration/scripts/single_use/utils/download_nltk_data.py
+++ /dev/null
@@ -1,53 +0,0 @@
-#!/usr/bin/env python3
-"""Download the NLTK corpora the plagiarism tools need.
-
-Extracted from `install_plagiarism_tools.sh`, which used to run this as an
-inline `python3 -c` block. Inline Python is invisible to ruff, mypy, pylint,
-bandit and pytest, so it lives in its own module instead.
-
-`nltk` is a dependency of the plagiarism toolchain, not of this repo, so it is
-absent from `meta/requirements.txt` on purpose. The installer script pip-installs
-it immediately before invoking this module, and probes for it first.
-"""
-
-from __future__ import annotations
-
-import importlib
-import sys
-
-#: Corpora required by the plagiarism checker. `punkt_tab` is the post-3.8.2
-#: replacement for `punkt`; both are listed so the script works on either.
-CORPORA = (
-    "punkt",
-    "stopwords",
-    "punkt_tab",
-    "averaged_perceptron_tagger",
-    "wordnet",
-)
-
-
-def main() -> int:
-    """Download every corpus, reporting any that fail.
-
-    Returns 1 if any download failed, so the calling shell script's `set -e`
-    sees a real failure instead of a silent partial install.
-    """
-    # Imported dynamically, not at module scope: nltk belongs to the plagiarism
-    # toolchain rather than this repo, so it is deliberately absent from
-    # meta/requirements.txt and unresolvable to static analysis. The installer
-    # pip-installs it immediately before invoking this module.
-    try:
-        nltk = importlib.import_module("nltk")
-    except ImportError:
-        sys.stderr.write("error: nltk is not installed\n")
-        return 1
-
-    failed = [corpus for corpus in CORPORA if not nltk.download(corpus)]
-    if failed:
-        sys.stderr.write(f"error: failed to download: {', '.join(failed)}\n")
-        return 1
-    return 0
-
-
-if __name__ == "__main__":
-    raise SystemExit(main())
diff --git a/linux_configuration/scripts/single_use/utils/install_plagiarism_tools.sh b/linux_configuration/scripts/single_use/utils/install_plagiarism_tools.sh
deleted file mode 100755
index 05deda52..00000000
--- a/linux_configuration/scripts/single_use/utils/install_plagiarism_tools.sh
+++ /dev/null
@@ -1,94 +0,0 @@
-#!/usr/bin/env bash
-# Install Free & Open Source Plagiarism Detection Tools for Text
-# Suitable for academic work (theses, papers, etc.)
-#
-# Tools installed:
-# 1. Python NLP-based similarity detection (sklearn, NLTK, spaCy)
-# 2. Sherlock text plagiarism detector
-# 3. Ferret (Java-based, if Java available)
-# 4. Optional: WCopyfind via Wine (Windows tool)
-
-set -euo pipefail
-
-INSTALL_DIR="${HOME}/.local/share/plagiarism-tools"
-VENV_DIR="${HOME}/.local/share/plagiarism-venv"
-
-# Colors for output
-RED='\033[0;31m'
-GREEN='\033[0;32m'
-YELLOW='\033[1;33m'
-NC='\033[0m' # No Color
-
-success() { echo -e "${GREEN}✓ $1${NC}"; }
-warn() { echo -e "${YELLOW}⚠ $1${NC}"; }
-error() { echo -e "${RED}✗ $1${NC}"; }
-
-SCRIPT_DIR="$(dirname "$(readlink -f "${BASH_SOURCE[0]}")")"
-# shellcheck source=lib/plagiarism_python.sh
-source "$SCRIPT_DIR/lib/plagiarism_python.sh"
-# shellcheck source=lib/plagiarism_sherlock.sh
-source "$SCRIPT_DIR/lib/plagiarism_sherlock.sh"
-# shellcheck source=lib/plagiarism_optional.sh
-source "$SCRIPT_DIR/lib/plagiarism_optional.sh"
-
-print_summary() {
-	echo ""
-	echo "=============================================="
-	echo " Installation Complete!"
-	echo "=============================================="
-	echo ""
-	echo "Installed tools:"
-	echo ""
-	echo "1. Python NLP Plagiarism Checker (TF-IDF, cosine similarity)"
-	echo "   Usage: plagcheck file1.txt file2.txt"
-	echo "          plagcheck --dir /path/to/documents/ --detailed"
-	echo "   Location: $INSTALL_DIR/check_plagiarism.py"
-	echo ""
-	echo "2. Sherlock (n-gram fingerprinting)"
-	echo "   Location: $INSTALL_DIR/sherlock/sherlock.py"
-	echo ""
-	echo "3. Python virtual environment with NLP libraries:"
-	echo "   - scikit-learn (TF-IDF, cosine similarity)"
-	echo "   - NLTK (tokenization, stopwords)"
-	echo "   - spaCy (NLP processing)"
-	echo "   - gensim (document similarity)"
-	echo "   - textdistance, fuzzywuzzy (string matching)"
-	echo "   Activate with: source $VENV_DIR/bin/activate"
-	echo ""
-	echo "Quick Start:"
-	echo "  plagcheck thesis_v1.pdf thesis_v2.pdf --detailed"
-	echo "  plagcheck --dir ./student_papers/ --threshold 0.4"
-	echo ""
-	echo "Note: Ensure ~/.local/bin is in your PATH:"
-	# Printed for the user to paste into their shell rc: $HOME and $PATH must
-	# stay literal rather than resolve to this machine's values.
-	# shellcheck disable=SC2016
-	echo '  export PATH="$HOME/.local/bin:$PATH"'
-	echo ""
-	echo "=============================================="
-
-	if [[ ":$PATH:" != *":$HOME/.local/bin:"* ]]; then
-		warn "Add ~/.local/bin to your PATH by adding this to ~/.bashrc or ~/.zshrc:"
-		# shellcheck disable=SC2016
-		echo '  export PATH="$HOME/.local/bin:$PATH"'
-	fi
-}
-
-main() {
-	echo "=============================================="
-	echo " Open Source Plagiarism Detection Installer"
-	echo " For Academic Text (Theses, Papers, etc.)"
-	echo "=============================================="
-	echo ""
-
-	mkdir -p "$INSTALL_DIR"
-
-	install_python_nlp_tools "$SCRIPT_DIR"
-	install_sherlock "$SCRIPT_DIR"
-	check_ferret
-	check_wcopyfind
-
-	print_summary
-}
-
-main "$@"
diff --git a/linux_configuration/scripts/single_use/utils/lib/plagiarism_optional.sh b/linux_configuration/scripts/single_use/utils/lib/plagiarism_optional.sh
deleted file mode 100755
index d6db2d4f..00000000
--- a/linux_configuration/scripts/single_use/utils/lib/plagiarism_optional.sh
+++ /dev/null
@@ -1,48 +0,0 @@
-#!/bin/bash
-# Stages 3-4: optional plagiarism tools that need a manual download.
-#
-# Sourced by install_plagiarism_tools.sh; split out to keep it under the
-# 250-line cap. Sourced rather than run, so it inherits the caller's strict
-# mode and the color/log helpers and INSTALL_DIR global defined above the
-# source line.
-
-# Ferret (Java-based) - Optional
-check_ferret() {
-	echo ""
-	echo "=== 3. Checking for Ferret (Java-based plagiarism tool) ==="
-
-	if ! command -v java &>/dev/null; then
-		warn "Java not installed, skipping Ferret"
-		return
-	fi
-
-	local ferret_dir="$INSTALL_DIR/ferret"
-	if [ -d "$ferret_dir" ]; then
-		return
-	fi
-
-	mkdir -p "$ferret_dir"
-	echo "Ferret is a Java-based tool from University of Hertfordshire."
-	echo "Download manually from: https://homepages.herts.ac.uk/~comqcln/Ferret/"
-	echo "Place JAR file in: $ferret_dir"
-	warn "Ferret requires manual download (academic license)"
-}
-
-# WCopyfind via Wine (Optional)
-check_wcopyfind() {
-	echo ""
-	echo "=== 4. WCopyfind Information (Windows tool, needs Wine) ==="
-
-	if command -v wine &>/dev/null; then
-		echo "Wine is available. WCopyfind can be run via Wine."
-		echo "Download from: https://plagiarism.bloomfieldmedia.com/software/wcopyfind/"
-		echo "Run with: wine /path/to/WCopyfind.exe"
-		warn "WCopyfind requires manual download"
-		return
-	fi
-
-	echo "Wine not installed. To use WCopyfind:"
-	echo "  1. Install wine: sudo apt install wine  (or equivalent)"
-	echo "  2. Download WCopyfind from: https://plagiarism.bloomfieldmedia.com/software/wcopyfind/"
-	warn "WCopyfind skipped (Wine not available)"
-}
diff --git a/linux_configuration/scripts/single_use/utils/lib/plagiarism_python.sh b/linux_configuration/scripts/single_use/utils/lib/plagiarism_python.sh
deleted file mode 100755
index 0eda6da7..00000000
--- a/linux_configuration/scripts/single_use/utils/lib/plagiarism_python.sh
+++ /dev/null
@@ -1,80 +0,0 @@
-#!/bin/bash
-# Stage 1: Python NLP-based plagiarism detection environment.
-#
-# Sourced by install_plagiarism_tools.sh; split out to keep it under the
-# 250-line cap. Sourced rather than run, so it inherits the caller's strict
-# mode and the color/log helpers and INSTALL_DIR/VENV_DIR globals defined
-# above the source line.
-
-# Create the venv, pip-install the NLP stack, and install check_plagiarism.py
-# (plus the plagiarism_similarity.py module it imports) and its "plagcheck"
-# wrapper. $1 is the directory install_plagiarism_tools.sh lives in, so both
-# scripts can be copied from beside it.
-install_python_nlp_tools() {
-	local script_dir="$1"
-
-	echo ""
-	echo "=== 1. Installing Python NLP-based Plagiarism Tools ==="
-
-	if ! command -v python3 &>/dev/null; then
-		error "Python 3 is required but not installed."
-		exit 1
-	fi
-
-	if [ ! -d "$VENV_DIR" ]; then
-		echo "Creating Python virtual environment..."
-		python3 -m venv "$VENV_DIR"
-		success "Virtual environment created at $VENV_DIR"
-	else
-		warn "Virtual environment already exists at $VENV_DIR"
-	fi
-
-	# Created at runtime by `python -m venv` above, so there is nothing on disk
-	# for the linter to follow at check time.
-	# shellcheck disable=SC1091
-	source "$VENV_DIR/bin/activate"
-
-	echo "Installing Python packages for text similarity detection..."
-	pip install --upgrade pip
-
-	pip install --progress-bar on \
-		scikit-learn \
-		nltk \
-		spacy \
-		gensim \
-		numpy \
-		pandas \
-		python-docx \
-		PyPDF2 \
-		beautifulsoup4 \
-		lxml \
-		textdistance \
-		fuzzywuzzy \
-		python-Levenshtein
-
-	success "Python NLP packages installed"
-
-	echo "Downloading NLTK data (stopwords, punkt tokenizer)..."
-	python3 "$script_dir/download_nltk_data.py"
-	success "NLTK data downloaded"
-
-	echo "Downloading spaCy English model..."
-	python3 -m spacy download en_core_web_sm 2>/dev/null || warn "spaCy model download may need manual install: python -m spacy download en_core_web_sm"
-	success "spaCy model installed"
-
-	install -m 755 "$script_dir/check_plagiarism.py" "$INSTALL_DIR/check_plagiarism.py"
-	install -m 644 "$script_dir/plagiarism_similarity.py" "$INSTALL_DIR/plagiarism_similarity.py"
-	success "Created plagiarism checker script at $INSTALL_DIR/check_plagiarism.py"
-
-	mkdir -p "$HOME/.local/bin"
-	cat >"$HOME/.local/bin/plagcheck" <<WRAPEOF
-#!/usr/bin/env bash
-# Wrapper for plagiarism checker
-source "$VENV_DIR/bin/activate"
-python "$INSTALL_DIR/check_plagiarism.py" "\$@"
-WRAPEOF
-	chmod +x "$HOME/.local/bin/plagcheck"
-	success "Created 'plagcheck' command in ~/.local/bin/"
-
-	deactivate
-}
diff --git a/linux_configuration/scripts/single_use/utils/lib/plagiarism_sherlock.sh b/linux_configuration/scripts/single_use/utils/lib/plagiarism_sherlock.sh
deleted file mode 100755
index a1dbd8aa..00000000
--- a/linux_configuration/scripts/single_use/utils/lib/plagiarism_sherlock.sh
+++ /dev/null
@@ -1,40 +0,0 @@
-#!/bin/bash
-# Stage 2: Sherlock text plagiarism detector.
-#
-# Sourced by install_plagiarism_tools.sh; split out to keep it under the
-# 250-line cap. Sourced rather than run, so it inherits the caller's strict
-# mode and the color/log helpers and INSTALL_DIR global defined above the
-# source line.
-
-# Clone the upstream sherlock-py tool, falling back to the bundled
-# sherlock.py (n-gram fingerprinting) when git or the clone is unavailable.
-# $1 is the directory install_plagiarism_tools.sh lives in, so the fallback
-# sherlock.py can be copied from beside it.
-install_sherlock() {
-	local script_dir="$1"
-
-	echo ""
-	echo "=== 2. Installing Sherlock Text Plagiarism Detector ==="
-
-	local sherlock_dir="$INSTALL_DIR/sherlock"
-	if [ -d "$sherlock_dir" ]; then
-		warn "Sherlock already installed at $sherlock_dir"
-		return
-	fi
-
-	if ! command -v git &>/dev/null; then
-		warn "Git not available, skipping Sherlock installation"
-		return
-	fi
-
-	# There are several Sherlock implementations; using a popular Python one
-	if git clone --depth 1 https://github.com/Zedeldi/sherlock-py.git "$sherlock_dir" 2>/dev/null; then
-		success "Sherlock installed at $sherlock_dir"
-		return
-	fi
-
-	warn "Could not clone sherlock-py, trying alternative..."
-	mkdir -p "$sherlock_dir"
-	install -m 755 "$script_dir/sherlock.py" "$sherlock_dir/sherlock.py"
-	success "Sherlock installed at $sherlock_dir"
-}
diff --git a/linux_configuration/scripts/single_use/utils/lib/tests/plagiarism_harness.sh b/linux_configuration/scripts/single_use/utils/lib/tests/plagiarism_harness.sh
deleted file mode 100755
index d57d633c..00000000
--- a/linux_configuration/scripts/single_use/utils/lib/tests/plagiarism_harness.sh
+++ /dev/null
@@ -1,130 +0,0 @@
-#!/usr/bin/env bash
-# lib/tests/plagiarism_harness.sh — fake python3/pip/git/java/wine behind the
-# install_plagiarism_tools.sh split.
-#
-# Sourced, not executed. Every external tool the installer calls is a shim on
-# PATH that records its invocation into $DEV and does the minimum real
-# filesystem work the calling code depends on (e.g. `python3 -m venv` must
-# leave a `bin/activate` a real `source` can run). INSTALL_DIR, VENV_DIR and
-# HOME all point into a throwaway tmpdir so nothing touches the real machine.
-set -euo pipefail
-
-SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
-UTILS_DIR="$(cd "${SCRIPT_DIR}/../.." && pwd)"
-
-PASS=0
-FAIL=0
-
-_t_pass() {
-	PASS=$((PASS + 1))
-	printf '  OK: %s\n' "$1"
-}
-
-_t_fail() {
-	FAIL=$((FAIL + 1))
-	printf '  FAIL: %s\n' "$1"
-}
-
-_t_eq() {
-	local want="$1" got="$2" what="$3"
-	if [[ "$got" == "$want" ]]; then
-		_t_pass "$what"
-	else
-		_t_fail "$what (want '${want}', got '${got}')"
-	fi
-}
-
-TEST_TMPDIR="$(mktemp -d)"
-trap 'rm -rf "${TEST_TMPDIR}"' EXIT
-
-readonly DEV="${TEST_TMPDIR}/device"
-readonly FAKE_BIN="${TEST_TMPDIR}/fake_bin"
-mkdir -p "${DEV}" "${FAKE_BIN}"
-
-# --- fake external tools ----------------------------------------------------
-
-cat >"${FAKE_BIN}/python3" <<'PYSHIM'
-#!/usr/bin/env bash
-set -euo pipefail
-DEV="${PLAGIARISM_TEST_DEV}"
-printf '%s\n' "python3 $*" >>"${DEV}/calls"
-if [[ "$1" == "-m" && "$2" == "venv" ]]; then
-	venv_dir="$3"
-	mkdir -p "${venv_dir}/bin"
-	cat >"${venv_dir}/bin/activate" <<'ACTEOF'
-deactivate() { :; }
-pip() { printf '%s\n' "pip $*" >>"${PLAGIARISM_TEST_DEV}/calls"; return 0; }
-ACTEOF
-	exit 0
-fi
-if [[ "$1" == "-m" && "$2" == "spacy" ]]; then
-	[[ -f "${DEV}/fail_spacy" ]] && exit 1
-	exit 0
-fi
-# python3 <script.py> [args...] — record which script ran and succeed.
-printf '%s\n' "ran:$1" >>"${DEV}/scripts_run"
-[[ -f "${DEV}/fail_$(basename "$1" .py)" ]] && exit 1
-exit 0
-PYSHIM
-chmod +x "${FAKE_BIN}/python3"
-
-cat >"${FAKE_BIN}/git" <<'GITSHIM'
-#!/usr/bin/env bash
-set -euo pipefail
-DEV="${PLAGIARISM_TEST_DEV}"
-printf '%s\n' "git $*" >>"${DEV}/calls"
-if [[ "$1" == "clone" ]]; then
-	[[ -f "${DEV}/fail_clone" ]] && exit 1
-	dest="${*: -1}"
-	mkdir -p "${dest}"
-	exit 0
-fi
-exit 0
-GITSHIM
-chmod +x "${FAKE_BIN}/git"
-
-cat >"${FAKE_BIN}/java" <<'JAVASHIM'
-#!/usr/bin/env bash
-exit 0
-JAVASHIM
-chmod +x "${FAKE_BIN}/java"
-
-cat >"${FAKE_BIN}/wine" <<'WINESHIM'
-#!/usr/bin/env bash
-exit 0
-WINESHIM
-chmod +x "${FAKE_BIN}/wine"
-
-export PLAGIARISM_TEST_DEV="${DEV}"
-export PATH="${FAKE_BIN}:${PATH}"
-
-# --- subject under test -----------------------------------------------------
-
-HOME="${TEST_TMPDIR}/home"
-INSTALL_DIR="${TEST_TMPDIR}/install"
-VENV_DIR="${TEST_TMPDIR}/venv"
-mkdir -p "${HOME}" "${INSTALL_DIR}"
-
-success() { :; }
-warn() { printf '%s\n' "$*" >>"${DEV}/warnings"; }
-error() { printf '%s\n' "$*" >>"${DEV}/errors"; }
-
-# shellcheck source=../plagiarism_python.sh
-. "${UTILS_DIR}/lib/plagiarism_python.sh"
-# shellcheck source=../plagiarism_sherlock.sh
-. "${UTILS_DIR}/lib/plagiarism_sherlock.sh"
-# shellcheck source=../plagiarism_optional.sh
-. "${UTILS_DIR}/lib/plagiarism_optional.sh"
-
-# reset_state — wipe INSTALL_DIR/VENV_DIR/DEV records between test groups so
-# each group starts from "nothing installed yet".
-reset_state() {
-	rm -rf "${INSTALL_DIR}" "${VENV_DIR}" "${DEV:?}/calls" "${DEV}/scripts_run" \
-		"${DEV}/warnings" "${DEV}/errors" 2>/dev/null || true
-	mkdir -p "${INSTALL_DIR}"
-	: >"${DEV}/calls"
-	: >"${DEV}/scripts_run"
-	: >"${DEV}/warnings"
-	: >"${DEV}/errors"
-}
-reset_state
diff --git a/linux_configuration/scripts/single_use/utils/lib/tests/test_install_plagiarism_tools.sh b/linux_configuration/scripts/single_use/utils/lib/tests/test_install_plagiarism_tools.sh
deleted file mode 100755
index a4af8f41..00000000
--- a/linux_configuration/scripts/single_use/utils/lib/tests/test_install_plagiarism_tools.sh
+++ /dev/null
@@ -1,114 +0,0 @@
-#!/usr/bin/env bash
-# End-to-end test for install_plagiarism_tools.sh's main()/print_summary(),
-# which stay in the entry script rather than a lib. Runs the real script as a
-# subprocess against the same fake python3/pip/git/java/wine as the lib tests,
-# with HOME pointed at a throwaway tmpdir so INSTALL_DIR/VENV_DIR (derived
-# from $HOME inside the script) land there too.
-set -euo pipefail
-
-SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
-
-# shellcheck source=plagiarism_harness.sh
-. "${SCRIPT_DIR}/plagiarism_harness.sh"
-
-readonly ENTRY_SCRIPT="${UTILS_DIR}/install_plagiarism_tools.sh"
-
-# The entry script derives INSTALL_DIR/VENV_DIR from $HOME itself (a plain
-# assignment near its top), independently of the lib tests' INSTALL_DIR/
-# VENV_DIR harness variables. Running it as a real subprocess against the
-# harness's $HOME means its paths land under $HOME, not under the harness's
-# separate INSTALL_DIR/VENV_DIR -- so assertions below re-derive the same
-# paths the script itself computes, rather than reusing the harness ones.
-readonly ENTRY_INSTALL_DIR="${HOME}/.local/share/plagiarism-tools"
-readonly ENTRY_VENV_DIR="${HOME}/.local/share/plagiarism-venv"
-
-# --- full run, PATH already in ~/.local/bin -------------------------------
-
-mkdir -p "${HOME}/.local/bin"
-OUT_FULL="${TEST_TMPDIR}/out_full.log"
-PATH="${FAKE_BIN}:${HOME}/.local/bin:${PATH}" bash "${ENTRY_SCRIPT}" >"${OUT_FULL}" 2>&1
-_status=$?
-
-if [[ "${_status}" -eq 0 ]]; then
-	_t_pass "install_plagiarism_tools.sh exits zero on a full successful run"
-else
-	_t_fail "install_plagiarism_tools.sh should exit zero (log: $(cat "${OUT_FULL}"))"
-fi
-
-if grep -q "Open Source Plagiarism Detection Installer" "${OUT_FULL}"; then
-	_t_pass "main prints the installer banner"
-else
-	_t_fail "main should print the installer banner"
-fi
-
-if grep -q "=== 1. Installing Python NLP-based Plagiarism Tools ===" "${OUT_FULL}"; then
-	_t_pass "main runs stage 1 (Python NLP tools)"
-else
-	_t_fail "main should run stage 1"
-fi
-
-if grep -q "=== 2. Installing Sherlock Text Plagiarism Detector ===" "${OUT_FULL}"; then
-	_t_pass "main runs stage 2 (Sherlock)"
-else
-	_t_fail "main should run stage 2"
-fi
-
-if grep -q "=== 3. Checking for Ferret" "${OUT_FULL}"; then
-	_t_pass "main runs stage 3 (Ferret)"
-else
-	_t_fail "main should run stage 3"
-fi
-
-if grep -q "=== 4. WCopyfind Information" "${OUT_FULL}"; then
-	_t_pass "main runs stage 4 (WCopyfind)"
-else
-	_t_fail "main should run stage 4"
-fi
-
-if grep -q "Installation Complete!" "${OUT_FULL}"; then
-	_t_pass "main prints the summary"
-else
-	_t_fail "main should print the summary"
-fi
-
-if grep -q "Location: ${ENTRY_INSTALL_DIR}/check_plagiarism.py" "${OUT_FULL}"; then
-	_t_pass "the summary reports the checker's installed location"
-else
-	_t_fail "the summary should report the checker's installed location"
-fi
-
-if grep -q "Location: ${ENTRY_INSTALL_DIR}/sherlock/sherlock.py" "${OUT_FULL}"; then
-	_t_pass "the summary reports sherlock's installed location"
-else
-	_t_fail "the summary should report sherlock's installed location"
-fi
-
-if grep -q "Activate with: source ${ENTRY_VENV_DIR}/bin/activate" "${OUT_FULL}"; then
-	_t_pass "the summary reports the venv activation command"
-else
-	_t_fail "the summary should report the venv activation command"
-fi
-
-# ~/.local/bin was already on PATH for this run, so the "add it to your PATH"
-# nudge must not repeat in the summary.
-if grep -q "Add ~/.local/bin to your PATH by adding this" "${OUT_FULL}"; then
-	_t_fail "the summary should not repeat the PATH nudge when ~/.local/bin is already on PATH"
-else
-	_t_pass "the summary skips the PATH nudge when ~/.local/bin is already on PATH"
-fi
-
-# --- ~/.local/bin missing from PATH: the summary nudges once --------------
-
-reset_state
-rm -rf "${HOME}/.local"
-OUT_NOPATH="${TEST_TMPDIR}/out_nopath.log"
-PATH="${FAKE_BIN}:${PATH}" bash "${ENTRY_SCRIPT}" >"${OUT_NOPATH}" 2>&1
-
-if grep -qc "Add ~/.local/bin to your PATH by adding this" "${OUT_NOPATH}"; then
-	_t_pass "the summary nudges to add ~/.local/bin to PATH when it is missing"
-else
-	_t_fail "the summary should nudge to add ~/.local/bin to PATH when it is missing"
-fi
-
-printf '\nResults: %d passed, %d failed\n' "${PASS}" "${FAIL}"
-[[ "${FAIL}" -eq 0 ]]
diff --git a/linux_configuration/scripts/single_use/utils/lib/tests/test_plagiarism_optional.sh b/linux_configuration/scripts/single_use/utils/lib/tests/test_plagiarism_optional.sh
deleted file mode 100755
index 87b703ff..00000000
--- a/linux_configuration/scripts/single_use/utils/lib/tests/test_plagiarism_optional.sh
+++ /dev/null
@@ -1,103 +0,0 @@
-#!/usr/bin/env bash
-# Unit tests for lib/plagiarism_optional.sh (Ferret/WCopyfind availability
-# checks) against the fake java/wine in plagiarism_harness.sh.
-set -euo pipefail
-
-SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
-
-# shellcheck source=plagiarism_harness.sh
-. "${SCRIPT_DIR}/plagiarism_harness.sh"
-
-_no_tool_path() {
-	local skip="$1"
-	local dir="${TEST_TMPDIR}/no_${skip}_bin"
-	mkdir -p "${dir}"
-	local tool
-	for tool in python3 git java wine cat chmod mkdir; do
-		[[ "${tool}" == "${skip}" ]] && continue
-		[[ -x "$(command -v "${tool}")" ]] && ln -sf "$(command -v "${tool}")" "${dir}/${tool}"
-	done
-	printf '%s' "${dir}"
-}
-
-# --- check_ferret: java available -----------------------------------------
-
-check_ferret >/tmp/plagiarism_optional_test.log 2>&1
-
-if [[ -d "${INSTALL_DIR}/ferret" ]]; then
-	_t_pass "check_ferret creates the ferret directory when java is available"
-else
-	_t_fail "check_ferret should create ${INSTALL_DIR}/ferret when java is available"
-fi
-
-if grep -q "requires manual download" "${DEV}/warnings"; then
-	_t_pass "check_ferret explains that Ferret needs a manual download"
-else
-	_t_fail "check_ferret should explain that Ferret needs a manual download"
-fi
-
-# check_ferret creates the directory as a marker; a second run must not
-# re-print the manual-download instructions once it already exists.
-: >"${DEV}/warnings"
-check_ferret >/tmp/plagiarism_optional_test.log 2>&1
-if grep -q "requires manual download" "${DEV}/warnings"; then
-	_t_fail "check_ferret should not repeat instructions once the ferret dir already exists"
-else
-	_t_pass "check_ferret is a no-op once the ferret directory already exists"
-fi
-
-# --- check_ferret: java unavailable ---------------------------------------
-
-reset_state
-no_java_path="$(_no_tool_path java)"
-PATH="${no_java_path}" check_ferret >/dev/null 2>&1
-
-if [[ -d "${INSTALL_DIR}/ferret" ]]; then
-	_t_fail "check_ferret should not create the ferret directory without java"
-else
-	_t_pass "check_ferret skips Ferret entirely when java is unavailable"
-fi
-
-if grep -q "Java not installed" "${DEV}/warnings"; then
-	_t_pass "check_ferret warns when java is unavailable"
-else
-	_t_fail "check_ferret should warn when java is unavailable"
-fi
-
-# --- check_wcopyfind: wine available ---------------------------------------
-
-: >"${DEV}/warnings"
-check_wcopyfind >/tmp/plagiarism_optional_test.log 2>&1
-
-if grep -q "Wine is available" /tmp/plagiarism_optional_test.log; then
-	_t_pass "check_wcopyfind reports Wine as available"
-else
-	_t_fail "check_wcopyfind should report Wine as available"
-fi
-
-if grep -q "requires manual download" "${DEV}/warnings"; then
-	_t_pass "check_wcopyfind warns that WCopyfind needs a manual download"
-else
-	_t_fail "check_wcopyfind should warn that WCopyfind needs a manual download"
-fi
-
-# --- check_wcopyfind: wine unavailable --------------------------------------
-
-: >"${DEV}/warnings"
-no_wine_path="$(_no_tool_path wine)"
-PATH="${no_wine_path}" check_wcopyfind >/tmp/plagiarism_optional_test.log 2>&1
-
-if grep -q "Wine not installed" /tmp/plagiarism_optional_test.log; then
-	_t_pass "check_wcopyfind reports Wine as unavailable"
-else
-	_t_fail "check_wcopyfind should report Wine as unavailable"
-fi
-
-if grep -q "skipped (Wine not available)" "${DEV}/warnings"; then
-	_t_pass "check_wcopyfind warns that WCopyfind was skipped"
-else
-	_t_fail "check_wcopyfind should warn that WCopyfind was skipped"
-fi
-
-printf '\nResults: %d passed, %d failed\n' "${PASS}" "${FAIL}"
-[[ "${FAIL}" -eq 0 ]]
diff --git a/linux_configuration/scripts/single_use/utils/lib/tests/test_plagiarism_python.sh b/linux_configuration/scripts/single_use/utils/lib/tests/test_plagiarism_python.sh
deleted file mode 100755
index e913e343..00000000
--- a/linux_configuration/scripts/single_use/utils/lib/tests/test_plagiarism_python.sh
+++ /dev/null
@@ -1,132 +0,0 @@
-#!/usr/bin/env bash
-# Unit tests for lib/plagiarism_python.sh against the fake python3/pip in
-# plagiarism_harness.sh. No real network or venv is needed.
-set -euo pipefail
-
-SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
-
-# shellcheck source=plagiarism_harness.sh
-. "${SCRIPT_DIR}/plagiarism_harness.sh"
-
-# --- happy path ---------------------------------------------------------
-
-install_python_nlp_tools "${UTILS_DIR}" >/tmp/plagiarism_python_test.log 2>&1
-_status=$?
-if [[ "${_status}" -eq 0 ]]; then
-	_t_pass "install_python_nlp_tools succeeds against a working toolchain"
-else
-	_t_fail "install_python_nlp_tools should succeed (log: $(cat /tmp/plagiarism_python_test.log))"
-fi
-
-if [[ -d "${VENV_DIR}" ]]; then
-	_t_pass "install_python_nlp_tools creates the virtual environment"
-else
-	_t_fail "install_python_nlp_tools should create ${VENV_DIR}"
-fi
-
-if grep -q '^python3 -m venv' "${DEV}/calls"; then
-	_t_pass "install_python_nlp_tools invokes python3 -m venv"
-else
-	_t_fail "install_python_nlp_tools should invoke python3 -m venv"
-fi
-
-if grep -q 'pip install --upgrade pip' "${DEV}/calls"; then
-	_t_pass "install_python_nlp_tools upgrades pip"
-else
-	_t_fail "install_python_nlp_tools should upgrade pip"
-fi
-
-if grep -q 'scikit-learn' "${DEV}/calls"; then
-	_t_pass "install_python_nlp_tools installs the NLP package set"
-else
-	_t_fail "install_python_nlp_tools should install scikit-learn and friends"
-fi
-
-if grep -q 'ran:.*download_nltk_data.py' "${DEV}/scripts_run"; then
-	_t_pass "install_python_nlp_tools runs download_nltk_data.py"
-else
-	_t_fail "install_python_nlp_tools should run download_nltk_data.py"
-fi
-
-if grep -q '^python3 -m spacy download' "${DEV}/calls"; then
-	_t_pass "install_python_nlp_tools downloads the spaCy model"
-else
-	_t_fail "install_python_nlp_tools should download the spaCy model"
-fi
-
-if [[ -x "${INSTALL_DIR}/check_plagiarism.py" ]]; then
-	_t_pass "install_python_nlp_tools installs check_plagiarism.py, executable"
-else
-	_t_fail "install_python_nlp_tools should install an executable check_plagiarism.py"
-fi
-
-if [[ -f "${INSTALL_DIR}/plagiarism_similarity.py" ]]; then
-	_t_pass "install_python_nlp_tools installs plagiarism_similarity.py alongside it"
-else
-	_t_fail "install_python_nlp_tools should install plagiarism_similarity.py -- \
-check_plagiarism.py imports it, so a missing copy breaks at runtime, not install time"
-fi
-
-if [[ -x "${HOME}/.local/bin/plagcheck" ]]; then
-	_t_pass "install_python_nlp_tools creates an executable plagcheck wrapper"
-else
-	_t_fail "install_python_nlp_tools should create an executable plagcheck wrapper"
-fi
-
-if grep -q "${VENV_DIR}/bin/activate" "${HOME}/.local/bin/plagcheck"; then
-	_t_pass "plagcheck wrapper activates the venv it was installed against"
-else
-	_t_fail "plagcheck wrapper should reference ${VENV_DIR}/bin/activate"
-fi
-
-if grep -q "${INSTALL_DIR}/check_plagiarism.py" "${HOME}/.local/bin/plagcheck"; then
-	_t_pass "plagcheck wrapper points at the installed checker script"
-else
-	_t_fail "plagcheck wrapper should point at ${INSTALL_DIR}/check_plagiarism.py"
-fi
-
-# --- venv already exists: warn, don't recreate --------------------------
-
-reset_state
-mkdir -p "${VENV_DIR}/bin"
-cat >"${VENV_DIR}/bin/activate" <<'ACTEOF'
-deactivate() { :; }
-pip() { printf '%s\n' "pip $*" >>"${PLAGIARISM_TEST_DEV}/calls"; return 0; }
-ACTEOF
-: >"${DEV}/calls"
-install_python_nlp_tools "${UTILS_DIR}" >/dev/null 2>&1
-if grep -q '^python3 -m venv' "${DEV}/calls"; then
-	_t_fail "install_python_nlp_tools should not recreate an existing venv"
-else
-	_t_pass "install_python_nlp_tools skips venv creation when one already exists"
-fi
-
-# --- python3 missing: fatal, not silently skipped ------------------------
-
-reset_state
-# A bare `rm -f "${FAKE_BIN}/python3"` does not hide python3 from
-# `command -v`: pyenv's shim is still reachable further down the real PATH.
-# Instead point PATH at a directory that shadows python3 with nothing before
-# falling through to the rest of the fakes, so `command -v python3` fails.
-NO_PYTHON_BIN="${TEST_TMPDIR}/no_python_bin"
-mkdir -p "${NO_PYTHON_BIN}"
-for tool in git java wine cat chmod mkdir; do
-	[[ -x "$(command -v "${tool}")" ]] && ln -sf "$(command -v "${tool}")" "${NO_PYTHON_BIN}/${tool}"
-done
-# install_python_nlp_tools calls `exit` (not `return`) on a missing python3,
-# by design -- a partial install with no interpreter must abort the whole
-# installer, not just this stage. Run it in a subshell so that exit only
-# ends the subshell, not this test script.
-if (PATH="${NO_PYTHON_BIN}" install_python_nlp_tools "${UTILS_DIR}" >/dev/null 2>&1); then
-	_t_fail "install_python_nlp_tools should exit non-zero without python3"
-else
-	_t_pass "install_python_nlp_tools exits non-zero when python3 is missing"
-fi
-if grep -q "Python 3 is required" "${DEV}/errors"; then
-	_t_pass "install_python_nlp_tools reports the missing-python3 error"
-else
-	_t_fail "install_python_nlp_tools should report the missing-python3 error"
-fi
-
-printf '\nResults: %d passed, %d failed\n' "${PASS}" "${FAIL}"
-[[ "${FAIL}" -eq 0 ]]
diff --git a/linux_configuration/scripts/single_use/utils/lib/tests/test_plagiarism_sherlock.sh b/linux_configuration/scripts/single_use/utils/lib/tests/test_plagiarism_sherlock.sh
deleted file mode 100755
index 5abaa0eb..00000000
--- a/linux_configuration/scripts/single_use/utils/lib/tests/test_plagiarism_sherlock.sh
+++ /dev/null
@@ -1,106 +0,0 @@
-#!/usr/bin/env bash
-# Unit tests for lib/plagiarism_sherlock.sh against the fake git in
-# plagiarism_harness.sh. No real network is needed.
-set -euo pipefail
-
-SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
-
-# shellcheck source=plagiarism_harness.sh
-. "${SCRIPT_DIR}/plagiarism_harness.sh"
-
-# --- upstream clone succeeds ---------------------------------------------
-
-install_sherlock "${UTILS_DIR}" >/dev/null 2>&1
-
-if grep -q '^git clone' "${DEV}/calls"; then
-	_t_pass "install_sherlock clones the upstream sherlock-py tool"
-else
-	_t_fail "install_sherlock should attempt a git clone"
-fi
-
-if [[ -d "${INSTALL_DIR}/sherlock" ]]; then
-	_t_pass "install_sherlock creates the sherlock directory on a successful clone"
-else
-	_t_fail "install_sherlock should create ${INSTALL_DIR}/sherlock"
-fi
-
-if [[ -f "${INSTALL_DIR}/sherlock/sherlock.py" ]]; then
-	_t_fail "a successful clone should not also install the bundled fallback sherlock.py"
-else
-	_t_pass "a successful clone does not install the bundled fallback sherlock.py"
-fi
-
-# --- already installed: no-op, warns -------------------------------------
-
-: >"${DEV}/calls"
-install_sherlock "${UTILS_DIR}" >/dev/null 2>&1
-
-if grep -q '^git clone' "${DEV}/calls"; then
-	_t_fail "install_sherlock should not re-clone when already installed"
-else
-	_t_pass "install_sherlock skips reinstalling when the directory already exists"
-fi
-
-if grep -q "already installed" "${DEV}/warnings"; then
-	_t_pass "install_sherlock warns that sherlock is already installed"
-else
-	_t_fail "install_sherlock should warn that sherlock is already installed"
-fi
-
-# --- clone fails: falls back to the bundled sherlock.py -------------------
-
-reset_state
-touch "${DEV}/fail_clone"
-install_sherlock "${UTILS_DIR}" >/dev/null 2>&1
-
-if [[ -f "${INSTALL_DIR}/sherlock/sherlock.py" ]]; then
-	_t_pass "install_sherlock falls back to the bundled sherlock.py when the clone fails"
-else
-	_t_fail "install_sherlock should fall back to the bundled sherlock.py when the clone fails"
-fi
-
-if [[ -x "${INSTALL_DIR}/sherlock/sherlock.py" ]]; then
-	_t_pass "the bundled fallback sherlock.py is installed executable"
-else
-	_t_fail "the bundled fallback sherlock.py should be installed executable"
-fi
-
-if grep -q "Could not clone sherlock-py" "${DEV}/warnings"; then
-	_t_pass "install_sherlock warns when the clone fails, before falling back"
-else
-	_t_fail "install_sherlock should warn when the clone fails"
-fi
-
-# The installed fallback must be byte-identical to the source file: it is the
-# one artifact the git-unavailable path can serve, so a stale copy would ship
-# broken plagiarism detection silently.
-if diff -q "${UTILS_DIR}/sherlock.py" "${INSTALL_DIR}/sherlock/sherlock.py" >/dev/null; then
-	_t_pass "the installed fallback sherlock.py matches the source file byte-for-byte"
-else
-	_t_fail "the installed fallback sherlock.py should match ${UTILS_DIR}/sherlock.py"
-fi
-
-# --- git not available: warns, does not attempt to install ----------------
-
-reset_state
-NO_GIT_BIN="${TEST_TMPDIR}/no_git_bin"
-mkdir -p "${NO_GIT_BIN}"
-for tool in python3 java wine cat chmod mkdir; do
-	[[ -x "$(command -v "${tool}")" ]] && ln -sf "$(command -v "${tool}")" "${NO_GIT_BIN}/${tool}"
-done
-PATH="${NO_GIT_BIN}" install_sherlock "${UTILS_DIR}" >/dev/null 2>&1
-
-if [[ -d "${INSTALL_DIR}/sherlock" ]]; then
-	_t_fail "install_sherlock should not create the sherlock directory without git"
-else
-	_t_pass "install_sherlock skips installation entirely when git is unavailable"
-fi
-
-if grep -q "Git not available" "${DEV}/warnings"; then
-	_t_pass "install_sherlock warns when git is unavailable"
-else
-	_t_fail "install_sherlock should warn when git is unavailable"
-fi
-
-printf '\nResults: %d passed, %d failed\n' "${PASS}" "${FAIL}"
-[[ "${FAIL}" -eq 0 ]]
diff --git a/linux_configuration/scripts/single_use/utils/plagiarism_similarity.py b/linux_configuration/scripts/single_use/utils/plagiarism_similarity.py
deleted file mode 100755
index b11bee85..00000000
--- a/linux_configuration/scripts/single_use/utils/plagiarism_similarity.py
+++ /dev/null
@@ -1,141 +0,0 @@
-#!/usr/bin/env python3
-"""Document reading and TF-IDF similarity scoring for check_plagiarism.py.
-
-Split out of `check_plagiarism.py` to keep it under the 250-line cap: this
-module holds the comparison engine (file reading, text preprocessing, and the
-TF-IDF/cosine-similarity computations), `check_plagiarism.py` keeps the CLI
-and reporting.
-
-Its third-party imports (numpy, sklearn, nltk) are the plagiarism toolchain's
-dependencies, not this repo's, so they are absent from `meta/requirements.txt`
-on purpose. Importing them dynamically (rather than with a static `import
-nltk`) keeps this module importable -- and lintable -- in an environment
-that doesn't have them installed, matching `download_nltk_data.py`'s
-extraction.
-"""
-
-from __future__ import annotations
-
-import importlib
-from pathlib import Path
-import sys
-
-#: Cosine similarity above this, for a single sentence pair, is worth reporting
-#: as a suspicious passage in --detailed mode.
-PASSAGE_SIMILARITY_THRESHOLD = 0.5
-
-nltk = importlib.import_module("nltk")
-stopwords = importlib.import_module("nltk.corpus").stopwords
-nltk_tokenize = importlib.import_module("nltk.tokenize")
-sent_tokenize = nltk_tokenize.sent_tokenize
-word_tokenize = nltk_tokenize.word_tokenize
-TfidfVectorizer = importlib.import_module(
-    "sklearn.feature_extraction.text"
-).TfidfVectorizer
-cosine_similarity = importlib.import_module(
-    "sklearn.metrics.pairwise"
-).cosine_similarity
-
-try:
-    stopwords.words("english")
-except LookupError:
-    nltk.download("stopwords", quiet=True)
-    nltk.download("punkt", quiet=True)
-
-
-def _out(line: str = "") -> None:
-    sys.stdout.write(line + "\n")
-
-
-def _read_pdf(filepath: str) -> str:
-    try:
-        pypdf2 = importlib.import_module("PyPDF2")
-    except ImportError:
-        _out("Warning: PyPDF2 not installed, cannot read PDF files")
-        return ""
-    reader = pypdf2.PdfReader(filepath)
-    return " ".join(page.extract_text() or "" for page in reader.pages)
-
-
-def _read_docx(filepath: str) -> str:
-    try:
-        docx = importlib.import_module("docx")
-    except ImportError:
-        _out("Warning: python-docx not installed, cannot read DOCX files")
-        return ""
-    doc = docx.Document(filepath)
-    return " ".join(para.text for para in doc.paragraphs)
-
-
-def read_file(filepath: str) -> str:
-    """Read text from a .pdf, .docx, or plain-text file."""
-    path = Path(filepath)
-    suffix = path.suffix.lower()
-
-    if suffix == ".pdf":
-        return _read_pdf(filepath)
-    if suffix == ".docx":
-        return _read_docx(filepath)
-    with path.open(encoding="utf-8", errors="ignore") as f:
-        return f.read()
-
-
-def preprocess_text(text: str) -> str:
-    """Lowercase, tokenize, and strip stopwords/non-alphanumerics."""
-    text = text.lower()
-    try:
-        stop_words = set(stopwords.words("english"))
-        words = word_tokenize(text)
-        words = [w for w in words if w.isalnum() and w not in stop_words]
-        return " ".join(words)
-    except LookupError:
-        return " ".join(text.split())
-
-
-def compute_similarity_matrix(documents: list[str]) -> list[list[float]]:
-    """Compute the TF-IDF cosine similarity matrix for a set of documents.
-
-    Returns a plain nested list, not the numpy array `cosine_similarity`
-    produces: numpy is a plagiarism-toolchain dependency the installer
-    pip-installs at runtime, deliberately absent from `meta/requirements.txt`,
-    so mypy cannot follow it into this repo's type-checked surface
-    (`no-any-unimported`). Converting at this boundary keeps every other
-    signature in the module concretely typed.
-    """
-    vectorizer = TfidfVectorizer(ngram_range=(1, 3), min_df=1, max_df=0.95)
-    tfidf_matrix = vectorizer.fit_transform(documents)
-    matrix: list[list[float]] = cosine_similarity(tfidf_matrix).tolist()
-    return matrix
-
-
-def find_similar_passages(
-    text1: str, text2: str, min_words: int = 5
-) -> list[tuple[str, str, float]]:
-    """Find similar sentence-level passages between two texts."""
-    sentences1 = [s for s in sent_tokenize(text1) if len(s.split()) >= min_words]
-    sentences2 = [s for s in sent_tokenize(text2) if len(s.split()) >= min_words]
-
-    if not sentences1 or not sentences2:
-        return []
-
-    all_sentences = sentences1 + sentences2
-    preprocessed = [preprocess_text(s) for s in all_sentences]
-
-    try:
-        vectorizer = TfidfVectorizer(ngram_range=(1, 2))
-        tfidf_matrix = vectorizer.fit_transform(preprocessed)
-    except ValueError:
-        return []
-
-    n1 = len(sentences1)
-    similarities = []
-
-    for i, s1 in enumerate(sentences1):
-        for j, s2 in enumerate(sentences2):
-            sim = cosine_similarity(
-                tfidf_matrix[i : i + 1], tfidf_matrix[n1 + j : n1 + j + 1]
-            )[0][0]
-            if sim > PASSAGE_SIMILARITY_THRESHOLD:
-                similarities.append((s1, s2, sim))
-
-    return sorted(similarities, key=lambda x: x[2], reverse=True)
diff --git a/linux_configuration/scripts/single_use/utils/sherlock.py b/linux_configuration/scripts/single_use/utils/sherlock.py
deleted file mode 100755
index 0e9460d6..00000000
--- a/linux_configuration/scripts/single_use/utils/sherlock.py
+++ /dev/null
@@ -1,108 +0,0 @@
-#!/usr/bin/env python3
-"""Sherlock - simple text plagiarism detector using n-gram fingerprinting.
-
-Extracted from `install_plagiarism_tools.sh`'s fallback path, which used to
-write this out via a `cat <<'SHERLOCKEOF'` heredoc when cloning the real
-sherlock-py failed. Inline heredoc Python is invisible to ruff, mypy, pylint,
-bandit and pytest, so it lives in its own module instead, matching
-`download_nltk_data.py`'s extraction.
-
-The installer copies this file into `$SHERLOCK_DIR/sherlock.py` only when
-`git clone` of the upstream tool fails.
-"""
-
-from __future__ import annotations
-
-import argparse
-import hashlib
-from itertools import combinations
-from pathlib import Path
-import sys
-
-#: Minimum number of input files for a pairwise comparison to make sense.
-MIN_FILES = 2
-
-
-def _out(line: str = "") -> None:
-    sys.stdout.write(line + "\n")
-
-
-def tokenize(text: str) -> list[str]:
-    """Split text into lowercase alphanumeric words."""
-    return [w.lower() for w in text.split() if w.isalnum()]
-
-
-def get_ngrams(tokens: list[str], n: int = 3) -> list[tuple[str, ...]]:
-    """Generate n-grams from a token list."""
-    return [tuple(tokens[i : i + n]) for i in range(len(tokens) - n + 1)]
-
-
-def fingerprint(text: str, ngram_size: int = 3, sample_rate: int = 4) -> set[str]:
-    """Create a document fingerprint from sampled n-gram hashes."""
-    tokens = tokenize(text)
-    ngrams = get_ngrams(tokens, ngram_size)
-
-    fingerprints = set()
-    for i, ng in enumerate(ngrams):
-        if i % sample_rate == 0:
-            h = hashlib.md5("".join(ng).encode(), usedforsecurity=False).hexdigest()[:8]
-            fingerprints.add(h)
-
-    return fingerprints
-
-
-def compare_documents(fp1: set[str], fp2: set[str]) -> float:
-    """Jaccard similarity between two fingerprints."""
-    if not fp1 or not fp2:
-        return 0.0
-    intersection = len(fp1 & fp2)
-    union = len(fp1 | fp2)
-    return intersection / union if union > 0 else 0.0
-
-
-def read_document(filepath: str) -> str:
-    """Read a document's raw text content."""
-    with Path(filepath).open(encoding="utf-8", errors="ignore") as f:
-        return f.read()
-
-
-def main() -> int:
-    """Fingerprint every input file and report pairwise similarity."""
-    parser = argparse.ArgumentParser(description="Sherlock - Text Plagiarism Detector")
-    parser.add_argument("files", nargs="+", help="Files to compare")
-    parser.add_argument(
-        "--ngram", "-n", type=int, default=3, help="N-gram size (default: 3)"
-    )
-    parser.add_argument(
-        "--threshold", "-t", type=float, default=0.1, help="Similarity threshold"
-    )
-
-    args = parser.parse_args()
-
-    if len(args.files) < MIN_FILES:
-        _out("Need at least 2 files to compare")
-        return 1
-
-    docs = {}
-    for f in args.files:
-        if Path(f).exists():
-            text = read_document(f)
-            docs[f] = fingerprint(text, args.ngram)
-
-    _out("\nSherlock Plagiarism Analysis")
-    _out("=" * 50)
-
-    files = list(docs.keys())
-    for file1, file2 in combinations(files, 2):
-        sim = compare_documents(docs[file1], docs[file2])
-        name1 = Path(file1).name
-        name2 = Path(file2).name
-        flag = " ⚠️ SUSPICIOUS" if sim >= args.threshold else ""
-        _out(f"{name1} <-> {name2}: {sim:.1%}{flag}")
-
-    _out("=" * 50)
-    return 0
-
-
-if __name__ == "__main__":
-    raise SystemExit(main())