create-pathfinder 1.3.0

package/skills/learn-feature/SKILL.md

@@ -0,0 +1,48 @@
+---
+name: learn-feature
+description: Generate a rich, interactive, feature-scoped lesson and quiz from completed implementation.
+---
+
+# Learn Feature
+
+Use after a feature is completed and accepted.
+
+## Read
+
+- completed feature spec and history entry
+- relevant diff/commits when available
+- only the implemented files and direct dependencies needed to explain the feature
+- tests and durable decisions
+
+## Output Format
+
+Use the project's learning preference. Default to self-contained HTML/CSS/JS. Use MDX only when the repository already supports it or the human selects it.
+
+Place output under:
+
+```text
+learning/features/[feature-slug]/
+```
+
+## Lesson Content
+
+- what changed and why
+- architecture and data/control flow
+- important files and responsibilities
+- key implementation decisions and tradeoffs
+- tests and verification
+- common mistakes and safe extension points
+- visual diagrams or interactive demonstrations when useful
+
+## Quiz Variety
+
+Include a useful mix when appropriate:
+
+- multiple choice
+- true/false with explanation
+- ordering or matching
+- predict-the-output/state
+- debugging scenario
+- short reflection or implementation challenge
+
+Provide immediate feedback and explanations. Keep the lesson scoped and avoid reading the whole repository.

package/skills/learning-review/SKILL.md

@@ -0,0 +1,76 @@
+---
+name: learning-review
+description: Review accumulated lessons and evidence, identify durable strengths and gaps, and create a focused reinforcement plan.
+argument-hint: feature|weekly|interview|concept
+---
+
+# Learning Review
+
+Use this skill periodically, not after every tiny change.
+
+The goal is to turn isolated feature lessons into a coherent learning path.
+
+## Read First
+
+1. `context/learning/learner-profile.md`
+2. `context/learning/progress.md`
+3. Recent lessons under `context/learning/lessons/`
+4. Recent completed feature specs and history
+5. Quiz or challenge evidence available in the repo
+
+## Review Modes
+
+### feature
+
+Review all learning evidence for one feature.
+
+### weekly
+
+Summarize recent concepts, repeated gaps, and one reinforcement plan.
+
+### interview
+
+Translate demonstrated project experience into interview-ready stories and identify claims that are not yet supported by evidence.
+
+### concept
+
+Review one concept across several features to determine whether understanding transfers.
+
+## Output
+
+Return:
+
+## Demonstrated Strengths
+
+Only claims supported by evidence.
+
+## Developing Areas
+
+Where understanding is partial or assisted.
+
+## Misconceptions or Missing Evidence
+
+Be direct and specific.
+
+## Connections Across Features
+
+Show how concepts reinforce each other.
+
+## Interview Stories
+
+For relevant work, produce concise STAR-style outlines without inventing business impact.
+
+## Next Three Learning Moves
+
+At most three, ordered by value.
+
+## Progress Changes
+
+Update `context/learning/progress.md` conservatively.
+
+## Rules
+
+- Do not equate code generated by an agent with knowledge demonstrated by the learner.
+- Do not create a giant curriculum.
+- Do not assign busywork.
+- Prefer reinforcement through upcoming product features.

package/skills/load-feature/SKILL.md

@@ -0,0 +1,17 @@
+---
+name: load-feature
+description: Prepare one feature for implementation by checking readiness, context size, dependencies, and delivery chunks.
+---
+
+# Load Feature
+
+1. Select the requested or next approved feature spec.
+2. Read its dependencies and the smallest relevant durable context.
+3. Check repository reality without scanning unrelated areas.
+4. Identify contradictions, missing decisions, stale paths, and prototype dependencies.
+5. Assess whether the feature fits a focused LLM context window.
+6. Split or revise it before implementation if the context is too broad.
+7. Populate `context/current-feature.md` with the feature, first delivery chunk, context boundary, assumptions, Git state, definition of done, and out-of-scope work.
+8. Present a short readiness summary.
+
+Do not implement, create Git history, or resolve `TBD` decisions silently.

package/skills/prototype/SKILL.md

@@ -0,0 +1,46 @@
+---
+name: prototype
+description: Create and iterate the cheapest useful artifact for validating a product, interaction, architecture, or technical assumption.
+---
+
+# Prototype
+
+Use this skill after debate recommends validation or when the human asks to see or test a direction before production planning.
+
+## Process
+
+1. Read the debate conclusion and relevant project context.
+2. Name the single primary assumption being validated.
+3. Recommend the cheapest useful prototype format.
+4. Define success/rejection criteria and an output location under `prototypes/`.
+5. Create only the prototype; avoid production architecture unless the task is explicitly a technical proof.
+6. Present focused review questions.
+7. On feedback, revise the same direction, replace it, approve it, or stop.
+8. Record approved direction, rejected assumptions, and explicit production exclusions in `context/project-overview.md`.
+
+## Supported Forms
+
+- wireframe or user-flow diagram
+- static visual concept
+- interactive HTML/CSS/JS
+- prototype using an already-installed project stack
+- API/integration/AI/animation/offline proof of concept
+- architecture or data-flow diagram
+
+Do not install a framework solely for a prototype without approval.
+
+## Human Review
+
+Ask:
+
+- Is the purpose or tested assumption clear?
+- Is the main flow/behavior correct?
+- What should be removed, revised, or replaced?
+- Is the direction approved for specification?
+
+## Rules
+
+- Prototype only what needs validation.
+- Label shortcuts and mocked behavior.
+- Do not silently promote prototype code into production.
+- Iterate rather than generating unrelated alternatives unless replacement is requested.

package/skills/quiz-me/SKILL.md

@@ -0,0 +1,78 @@
+---
+name: quiz-me
+description: Assess understanding of a recently taught feature using varied, evidence-based questions and targeted feedback.
+argument-hint: optional lesson path or feature name
+---
+
+# Quiz Me
+
+Use this skill after `teach-feature`, or when the user wants to test understanding of a feature or concept.
+
+The goal is retrieval and diagnosis, not entertainment and not a long exam.
+
+## Read First
+
+1. `context/learning/learner-profile.md`
+2. The relevant lesson in `context/learning/lessons/`
+3. `context/current-feature.md`
+4. Only the implementation or tests needed to verify answers
+5. `context/learning/progress.md`
+
+## Quiz Shape
+
+Create 5 questions by default. Mix at least three formats:
+
+- Predict what happens
+- Explain why a decision was made
+- Trace data or event flow
+- Find or diagnose a bug
+- Choose between architectural alternatives
+- Identify an accessibility or performance risk
+- Interpret a focused code excerpt
+- Explain how a test proves behavior
+
+Avoid:
+
+- pure trivia
+- questions answerable only by memorizing file names
+- five multiple-choice questions
+- ambiguous questions without a defensible answer
+- testing concepts not present in the lesson or code
+
+## Delivery
+
+Ask one question at a time.
+
+After each answer:
+
+1. Mark it as correct, partially correct, or incorrect.
+2. Explain the key point concisely.
+3. Ask one targeted follow-up only when it will diagnose a meaningful gap.
+4. Continue to the next question.
+
+Do not reveal all answers before the learner responds.
+
+## Scoring
+
+At the end, report:
+
+- Strong understanding
+- Partial understanding
+- Gaps to reinforce
+- One recommended next action
+
+Do not use a percentage unless every question had a clear scoring basis.
+
+## Progress Update
+
+Update `context/learning/progress.md` with demonstrated evidence.
+
+Use the confidence scale already defined there.
+
+A correct answer with heavy hints is `practiced`, not `independent`.
+
+## Scope
+
+- Do not change product code.
+- Do not inflate the quiz to cover the whole stack.
+- Do not treat one quiz as proof of mastery.