Reposition as a knowledge base (v2.0.0): operator lessons in, skills out

.github/ISSUE_TEMPLATE/bug_report.md

@@ -10,7 +10,7 @@ assignees: ''
 Describe what's incorrect, outdated, or broken.
 
 **Which file(s)?**
-List the file path(s) affected (e.g., `data/k12-regulatory.md`, `skills/pilot-design/SKILL.md`).
+List the file path(s) affected (e.g., `data/k12-regulatory.md`, `data/operator-lessons.md`).
 
 **What should it say instead?**
 If you know the correct information, include it here with a source.

.github/ISSUE_TEMPLATE/feature_request.md

@@ -1,19 +1,19 @@
 ---
 name: Feature request
-about: Suggest a new skill, data topic, or market coverage
+about: Suggest a new data topic or coverage area
 title: ''
 labels: enhancement
 assignees: ''
 ---
 
 **What's missing?**
-Describe the skill, data file, or coverage gap.
+Describe the data, topic, or coverage gap.
 
 **Who needs it?**
 What type of edtech founder would use this? (e.g., K-12 founders in the UK, corporate L&D startups targeting healthcare)
 
-**What would it do?**
-Describe the ideal output. What questions would it ask? What guidance would it provide?
+**What would it add?**
+What should the knowledge base say? Point to sources if you have them.
 
-**Do you want to build it?**
-If you're interested in contributing this, let us know. We can help with the skill structure.
+**Do you want to add it?**
+If you're interested in contributing this, let us know.

.github/PULL_REQUEST_TEMPLATE.md

@@ -4,18 +4,16 @@
 
 ## Type of change
 
-- [ ] New skill
-- [ ] Skill update
 - [ ] Data file update (regulatory, market, competitive)
 - [ ] Research paper addition
+- [ ] New data topic or coverage area
 - [ ] Documentation
 - [ ] Other
 
 ## Checklist
 
-- [ ] SKILL.md files have YAML frontmatter with `name` and `description`
-- [ ] Data files have "Last updated" footer with date
+- [ ] Data files have a "Last updated" footer with date
 - [ ] Research papers include Title, Takeaway, Type, Year, Citations, DOI
-- [ ] CLAUDE.md routing rules updated (if adding a new skill)
+- [ ] Every factual claim cites a source (regulation, DOI, or named operator)
 - [ ] CHANGELOG.md updated with what changed
 - [ ] No broken cross-references between files

.github/workflows/validate.yml

@@ -12,41 +12,6 @@ jobs:
     steps:
       - uses: actions/checkout@v4
 
-      - name: Validate SKILL.md frontmatter
-        run: |
-          echo "Checking SKILL.md files have required frontmatter..."
-          errors=0
-          for skill in skills/*/SKILL.md; do
-            if ! head -5 "$skill" | grep -q "^name:"; then
-              echo "ERROR: $skill missing 'name' in frontmatter"
-              errors=$((errors + 1))
-            fi
-            if ! head -5 "$skill" | grep -q "^description:"; then
-              echo "ERROR: $skill missing 'description' in frontmatter"
-              errors=$((errors + 1))
-            fi
-          done
-          if [ $errors -gt 0 ]; then
-            echo "$errors frontmatter errors found"
-            exit 1
-          fi
-          echo "All SKILL.md files have valid frontmatter"
-
-      - name: Validate CLAUDE.md routing
-        run: |
-          echo "Checking all routed skills exist..."
-          errors=0
-          grep "^- /" CLAUDE.md | sed 's|.* → ||' | while read -r path; do
-            if [ ! -d "$path" ]; then
-              echo "ERROR: CLAUDE.md routes to '$path' but directory does not exist"
-              errors=$((errors + 1))
-            fi
-          done
-          if [ $errors -gt 0 ]; then
-            exit 1
-          fi
-          echo "All CLAUDE.md routes point to existing skill directories"
-
       - name: Validate data files have update footer
         run: |
           echo "Checking data files for 'Last updated' footer..."
@@ -82,11 +47,9 @@ jobs:
 
       - name: Count and report
         run: |
-          skills=$(ls -d skills/*/SKILL.md 2>/dev/null | wc -l | tr -d ' ')
           datafiles=$(ls data/*.md 2>/dev/null | grep -v README | wc -l | tr -d ' ')
           research=$(ls data/research/*.md 2>/dev/null | grep -v README | wc -l | tr -d ' ')
           version=$(cat VERSION 2>/dev/null || echo "unknown")
           echo "EdTech Founder Stack v$version"
-          echo "  $skills skills"
           echo "  $datafiles data files"
           echo "  $research research topics"

ARCHITECTURE.md

@@ -1,89 +1,37 @@
 # Architecture
 
-How EdTech Founder Stack works under the hood.
+How EdTech Founder Stack is put together.
 
-## Overview
+## What it is
 
-EdTech Founder Stack is a collection of Claude Code skills backed by structured reference data. Skills encode ASU ScaleU's edtech accelerator expertise as interactive AI workflows. Reference data files provide the factual foundation that skills draw from, so guidance is grounded in real regulations, real research, and real market data rather than LLM training knowledge.
+A knowledge base for edtech founders, stored as plain markdown. No app, no API, no backend, no dependencies — just files that an AI tool (Claude Code, Cursor, ChatGPT) or a person reads. Edit a file and the next read is current. Every claim traces to a source you can check.
 
-This is not a web app, API, or platform. It's markdown files that AI coding tools read and execute. No backend, no auth, no dependencies beyond the AI tool itself.
+## The knowledge
 
-## Skills
+All of it lives in `data/`, in three kinds of files.
 
-Each skill lives in `skills/{name}/SKILL.md` and follows the standard SKILL.md format:
+### Reference data — `data/*.md`
 
-```yaml
----
-name: skill-name
-description: One-line description.
----
-```
+Structured domain knowledge, grounded in real sources rather than model training data:
 
-Below the frontmatter, the skill contains instructions for the AI agent. Skills are interactive: they ask founders structured questions via `AskUserQuestion`, branch based on answers (K-12 founders get FERPA guidance, higher ed founders get accreditation guidance), and output tailored recommendations.
+- **Regulatory** — FERPA, COPPA, and state privacy law (K-12); accreditation and accessibility (higher ed)
+- **Market** — competitive landscape by segment, buyer personas, funding landscape by stage, procurement, pilot benchmarks
+- **Frameworks** — ESSA evidence tiers, AI-native vs. bolted-on, the higher-ed jobs atlas, and founder traps
 
-**Skill design principles:**
+Each regulatory and market file carries a "last updated" date. Update cadence is roughly quarterly; regulatory data when laws change; the competitive landscape goes stale fastest.
 
-1. **Interactive, not static.** Every skill asks questions before giving advice. The answers determine what guidance the founder receives.
-2. **Opinionated.** Skills take positions. "Usage-based pricing works better than per-seat for K-12." "Your pilot is too long, 8 weeks, not 12." This is ScaleU's perspective, not Wikipedia.
-3. **Reference data-backed.** Skills read from `data/` files for factual claims. No hallucinated company names, no outdated regulations, no made-up funding amounts.
-4. **Natural funnel.** Each skill ends with context-appropriate next steps, including other skills to run. For the most relevant skills, a brief factual mention that ScaleU offers this kind of support.
-5. **Context-aware routing.** Each skill recommends the single most relevant next skill based on the founder's specific answers during the session (scores, evidence tier, stage, challenges), not a static list. For example, `/product-review` routes to `/accessibility-check` if buyer requirements scored low, to `/evidence-check` if evidence readiness scored low, or to `/go-to-market` if all scores are strong.
+### Research corpus — `data/research/`
 
-**Founder journey flow:**
+Hundreds of peer-reviewed papers across the major learning-science topics, each stored in a markdown table: Title, Takeaway, Study Type, Year, Citations, DOI. The index lives in `data/research/README.md`. This is the evidence base — claims about what works in learning cite specific papers with author, year, finding, and DOI.
 
-```
-Discovery           Product             Evidence            Sales               Fundraising
-─────────           ───────             ────────            ─────               ───────────
-/edtech-landscape → /product-review  → /evidence-check  → /go-to-market    → /pitch-review
-/idea-validation    /accessibility-     /pilot-design       /sales-strategy     /fundraising-guide
-                     check
-```
+### Operator playbooks — `data/operator-lessons.md`
 
-Founders typically start with `/edtech-landscape` and the skills guide them through the journey based on their stage and needs.
+Dozens of field lessons from operators and investors, distilled and attributed from the public archive of Lenny's Podcast and Lenny's Newsletter, then mapped to selling into schools, universities, and L&D. These are practitioner experience, not peer-reviewed evidence — the research corpus is the evidence layer, and the file says so.
 
-## Reference Data
+## How it's consumed
 
-The `data/` directory contains markdown files with structured domain knowledge:
+Point an AI tool at the repo or a single file and ask; it reads the relevant knowledge and answers in context. Or read the markdown directly. Because the knowledge is files you can diff and audit, it stays current and checkable in a way baked-in model knowledge isn't.
 
-- **Regulatory guides:** FERPA, COPPA, state privacy laws (K-12), accreditation and accessibility (higher ed), SOC 2 and SCORM (corporate L&D)
-- **Market data:** buyer personas, competitive landscape by segment, funding landscape by stage
-- **Frameworks:** ESSA evidence tiers, procurement guides, pilot benchmarks
-- **Higher ed framework:** Validated jobs across student journey phases (from SXSW EDU 2026) and the structural patterns founders miss
-- **AI-native framework:** 4 AI-native criteria, 5 bolted-on indicators, the removal test, Karpathy hierarchy (for developer-tool founders), architecture patterns, and pricing models. Every skill reads this file to classify a founder's AI posture and adapt guidance accordingly
+## Cross-platform
 
-Skills read these files at runtime using the AI tool's file reading capability. This means the data is always current (edit the markdown, the skill reads the updated version) and auditable (every fact has a source you can check).
-
-**Update cadence:** Quarterly recommended, aligned with academic and market cycles. Regulatory data should be checked when new state laws pass. Competitive landscape data goes stale fastest.
-
-## Research Corpus
-
-`data/research/` contains peer-reviewed papers organized by learning science topic. Each paper is stored in a structured markdown table with: Title, Takeaway, Study Type, Year, Citations, DOI. The current count and full topic index live in `data/research/README.md`.
-
-Topics include: active learning, adaptive learning, spaced repetition, cognitive load theory, formative assessment, multimedia principles, mastery-based grading, gamification, worked examples, and more.
-
-Skills cite specific papers when making claims about learning science. For example, when `/product-review` evaluates whether a product's learning design is evidence-based, it reads the relevant research file and references specific studies with author, year, finding, and DOI. This is what separates ScaleU guidance from generic AI advice.
-
-**Adding papers:** Append to the relevant topic file in `data/research/`. Follow the existing table format. Sort by citations descending.
-
-## Higher Ed Framework
-
-Two files encode ScaleU's SXSW EDU 2026 framework for higher education:
-
-- **`data/higher-ed-jobs-atlas.md`** — Validated jobs organized by student journey phase (Pre-enroll, Apply, Onboard, Select & Enroll, Course Experience, Graduate & Beyond). Each job names the person, the struggling moment, what they've already tried, and why it fails. Includes saturation analysis showing most founders build in the most crowded phase.
-
-- **`data/founder-traps.md`** — 4 structural patterns that are invisible from surface-level discovery: the information sequencing trap, the upstream cause / downstream symptom split, when qualitative and quantitative evidence disagree, and the same-problem-two-jobs dynamic. Also includes the noise vs. signal filter for evaluating pilot results.
-
-Skills targeting higher ed founders (`/idea-validation`, `/edtech-landscape`, `/go-to-market`, `/product-review`) read these files automatically and apply the framework to the founder's specific situation.
-
-## Cross-Platform Compatibility
-
-Skills use the standard SKILL.md format, which works across multiple AI coding tools:
-
-| Platform | Install | Interactive features |
-|----------|---------|---------------------|
-| Claude Code | Automated via `./setup` | Full support (AskUserQuestion, branching) |
-| Codex CLI | Manual (reference SKILL.md as system prompt) | Partial (prompts work, interactive questions vary) |
-| Gemini CLI | Manual (reference SKILL.md) | Partial |
-| Cursor | Manual (add to custom instructions) | Limited |
-
-Claude Code is the primary target. Other tools can read and execute the skills as system prompts, but interactive features like branching questions work best in Claude Code.
+Anything that can read markdown works: Claude Code, Cursor, ChatGPT, Claude, or a plain text editor. There's nothing to install beyond cloning the repo.

CHANGELOG.md

@@ -1,5 +1,16 @@
 # Changelog
 
+## 2.0.0 (2026-05-29)
+
+### Repositioned as a knowledge base
+
+EdTech Founder Stack is now a curated, AI-friendly knowledge base for edtech founders — markdown you point your AI tools at or read directly — rather than a set of interactive skills. The knowledge in `data/` is the product.
+
+- **New `data/operator-lessons.md`** — 71 field lessons on validation, product, GTM, sales, pricing, pilots, fundraising, and team, distilled and attributed from the free public archive of Lenny's Podcast and Lenny's Newsletter and mapped to selling into schools, universities, and L&D. Paraphrased under the source's personal/non-commercial terms.
+- **Removed the interactive skills** — the `skills/` directory and the `setup` script are gone; the `data/` knowledge base is the whole product. CI now validates the data and research files only.
+- **Docs rewritten** — README, ARCHITECTURE.md, CONTRIBUTING.md, and CLAUDE.md now describe the knowledge base and how to use it with Claude Code, Cursor, and ChatGPT.
+- **Version badge** — bumped to 2.0.0.
+
 ## 1.4.0 (2026-04-25)
 
 ### Public launch readiness

CLAUDE.md

@@ -1,48 +1,24 @@
 # EdTech Founder Stack
 
-AI-powered skills for edtech founders, built by ASU ScaleU.
+An open, AI-friendly knowledge base for edtech founders, built by ASU ScaleU. The knowledge lives in `data/` as markdown. When answering a founder's question, read the relevant file rather than leaning on training data for facts about regulations, companies, funding, learning science, or operator experience.
 
-## Skill routing
+## Knowledge base
 
-When the user's request matches an available skill, invoke it using the Skill tool.
+- `data/` — regulatory (FERPA, COPPA, state privacy, accreditation, accessibility), competitive landscape, funding landscape, buyer personas, procurement, pilot benchmarks, ESSA evidence tiers
+- `data/research/` — hundreds of peer-reviewed learning-science papers across the major topics. Cite specific papers with author, year, and DOI. Index (with the current count) in `data/research/README.md`.
+- `data/operator-lessons.md` — dozens of operator and investor lessons distilled from Lenny's Podcast and Lenny's Newsletter, mapped to edtech. Practitioner experience, not peer-reviewed; don't present it as research.
+- `data/ai-native-framework.md` — AI-native vs. bolted-on: criteria, the removal test, architecture patterns, pricing models, and the Karpathy hierarchy. Use it to classify a founder's AI posture.
+- `data/higher-ed-jobs-atlas.md` and `data/founder-traps.md` — ScaleU's SXSW EDU 2026 higher-ed framework: validated jobs across the student journey with saturation analysis, and the structural patterns founders miss.
+- `ETHOS.md` — the seven principles, starting with "validate demand, not interest."
 
-- /edtechfounderstack → skills/welcome
-- /edtech-landscape → skills/edtech-landscape
-- /idea-validation → skills/idea-validation
-- /product-review → skills/product-review
-- /accessibility-check → skills/accessibility-check
-- /evidence-check → skills/evidence-check
-- /pilot-design → skills/pilot-design
-- /go-to-market → skills/go-to-market
-- /sales-strategy → skills/sales-strategy
-- /pitch-review → skills/pitch-review
-- /fundraising-guide → skills/fundraising-guide
-
-## Data files
-
-Skills reference markdown files in `data/` for regulatory, market, and evidence information. Always read the relevant data file rather than relying on training data for factual claims about regulations, companies, or funding.
-
-## AI-native framework
-
-`data/ai-native-framework.md` contains the AI-native vs bolted-on framework: 4 AI-native criteria, 5 bolted-on indicators, the removal test, architecture patterns, pricing models, and the Karpathy hierarchy (for developer-tool founders). Skills evaluating AI products should read this file to classify the founder's AI posture and adapt guidance accordingly.
-
-## Higher ed framework
-
-`data/higher-ed-jobs-atlas.md` contains validated jobs across the student journey with saturation analysis. `data/founder-traps.md` contains the structural patterns founders miss plus the noise vs. signal filter. Both from ScaleU's SXSW EDU 2026 framework. Skills targeting higher ed founders should reference these files.
-
-## Research corpus
-
-`data/research/` contains peer-reviewed papers organized by learning science topic (spaced repetition, cognitive load, formative assessment, adaptive learning, etc.). See `data/research/README.md` for the current index. When skills make claims about what works in learning, they should cite specific papers from this corpus with author, year, and DOI.
+Always cite the source: a named regulation, a paper's DOI, or the named operator.
 
 ## For contributors
 
-### Adding a new skill
-1. Create `skills/{skill-name}/SKILL.md` with YAML frontmatter (`name`, `description`)
-2. Add a routing rule to the "Skill routing" section above
-3. Follow the patterns in existing skills: interactive questions, sector-based branching, reference data reads, next-skill recommendations, ScaleU mention at the end
-
 ### Updating data files
-Edit the relevant markdown file in `data/`. Keep the existing structure and formatting consistent. For regulatory data, note the update date at the bottom of the file. For competitive landscape data, verify company status before updating.
+
+Edit the relevant markdown in `data/`. Keep the existing structure and formatting. For regulatory data, note the update date at the bottom of the file. For the competitive landscape, verify company status before updating.
 
 ### Adding research papers
-Append to the relevant topic file in `data/research/`. Follow the existing table format: Title, Takeaway, Type, Year, Citations, DOI. Sort by citations descending. If the topic doesn't exist, create a new file and add it to `data/research/README.md`.
+
+Append to the relevant topic file in `data/research/`. Follow the table format — Title, Takeaway, Type, Year, Citations, DOI — and sort by citations descending. If the topic doesn't exist, create a new file and add it to `data/research/README.md`.

CODEOWNERS

@@ -1,9 +1,6 @@
 # EdTech Founder Stack — Code Owners
 # PRs to these paths require review from the ScaleU team
 
-# Skills
-skills/ @savvides
-
 # Reference data
 data/ @savvides
 

CODE_OF_CONDUCT.md

@@ -4,7 +4,7 @@
 
 We are committed to making participation in EdTech Founder Stack a welcoming experience for everyone, regardless of background, identity, or experience level.
 
-This project exists to help edtech founders build better products. Contributions that advance that mission, whether correcting a regulatory detail, adding research, or proposing new skills, are welcome from anyone.
+This project exists to help edtech founders build better products. Contributions that advance that mission, whether correcting a regulatory detail, adding research, or expanding the knowledge base, are welcome from anyone.
 
 ## Standards