Day 14 was an unplanned but highly productive housekeeping sprint. What started as
a planned day off turned into a deep dive into site integrity tooling. Three new bash
scripts were built, tested, and deployed to /usr/local/bin/ — a complete
link audit toolkit for the AI Education Project site. The scripts evolved iteratively
through real output, fixing edge cases as live data revealed them.
The session also produced a formal Housekeeping Work Order HTML document
categorizing all 104 orphaned files into actionable buckets. By end of session the orphan
count was reduced from 104 to 47 through script refinement (excluding SAVE*
and pre-refactor/ archive trees) and actual file cleanup work.
Single-directory landing page validator. Checks hrefs in an index.html
against files in that same directory. Reports LINKED, ORPHAN, MISSING, and SYMLINK status.
Handles index.html symlinks correctly — follows the symlink, reads the real
file, skips the real file from orphan reporting.
index.html and shows target in headerRecursive tree-wide link auditor. Scans ALL index.html files in the tree,
resolves every local href to an absolute path, and reports which files are linked anywhere
in the tree vs. truly orphaned. Four iterative fixes were applied during the session:
href="../../TMA/AI/" now marks ALL files in that dir as LINKEDSAVED/ backup directories ignored[ 4/4 ] Symlinks Found section — symlinks never inflate orphan countFormal HTML work order generated from the first full check_ai_tree.sh run.
104 orphaned files categorized into: Fix Links, Safe to Delete, Archive, Add Links, and
Review buckets. Styled in project color scheme. Lives in
collaboration/building-in-public/.
| Run | Orphans | Broken Links | Change Made |
|---|---|---|---|
| Run 1 | 104 | 6 | Baseline |
| Run 2 | 73 | 4 | Symlink fix applied |
| Run 3 | 69 | 4 | SAVE* exclusion added |
| Run 4 | 69 | 2 | Directory href resolution fixed |
| Run 5 | 47 | 1 | pre-refactor exclusion added |
collaboration/index.html → fix ../project-diary.html to project-diary.htmlAssess_Your_AI_Knowledge.html → link from root index.htmlcollaboration/building-this-site.html → link from collaboration/index.htmldeep-dive/advanced-topics.html → link from deep-dive/index.htmllearning/introduction-to-ai.html → link from learning/index.htmllearning/Recommended_Learning_Path-AI-LLM.html → link from learning/index.htmlcollaboration/building-in-public/*.html (13 diary entries) → link from collaboration/project-diary.htmlFILE-MIGRATION-PLAN.txtlearning/AI-Learing-Plan-For-Beginners.html — typo filename, superseded by correct versioncollaboration/addendum/STATUS-REPORT-DAY1-3.txt — duplicate of .html versioncollaboration/addendum/*.txt (20 daily WWNTD notes) — move out of docroot or link from diary entriesresources/ dir — needs index.html or consolidate into learning/AI-Cat_use_List_ai-x.html, AI_learning_plan_ai-x.html, Recommended_Learning_Path-AI-LLM.htmlThe iterative script development process today was a good demonstration of the AI collaboration model in practice. Real output from a live site revealed edge cases (symlinks, directory hrefs, archive trees) that no amount of upfront design would have caught. Each run of the script produced data that drove the next improvement. Five iterations, live data, working tool.
The “day off” that wasn’t. 40 years in and the pull of a good problem is still stronger than the TV remote.
cmh — AI Education Project Day 14 — March 8, 2026
Next session: Day 15 — Work off the remaining orphan list, then back to deep-dive/ CLI tools pages.