Polish the v0.1.0 launch experience

.github/ISSUE_TEMPLATE/bug_report.yml

@@ -0,0 +1,49 @@
+name: Bug report
+description: Report a reproducible problem in Skill Feedback Engine.
+title: "[Bug]: "
+labels:
+  - bug
+body:
+  - type: textarea
+    id: description
+    attributes:
+      label: What happened?
+      description: Describe the behavior and its impact.
+    validations:
+      required: true
+  - type: textarea
+    id: reproduction
+    attributes:
+      label: How can we reproduce it?
+      description: Include the smallest safe command sequence or fixture that demonstrates the issue.
+      placeholder: |
+        1. Run ...
+        2. Observe ...
+    validations:
+      required: true
+  - type: textarea
+    id: expected
+    attributes:
+      label: What did you expect?
+    validations:
+      required: true
+  - type: textarea
+    id: environment
+    attributes:
+      label: Environment
+      description: Include the release or commit, Python version, operating system, and installation method.
+      placeholder: |
+        Skill Feedback Engine: v0.1.0
+        Python: 3.12
+        OS: macOS 15
+        Install: pip editable checkout
+    validations:
+      required: true
+  - type: checkboxes
+    id: privacy
+    attributes:
+      label: Privacy check
+      description: Observations and evidence may contain private work context.
+      options:
+        - label: I removed secrets, personal information, raw conversation history, and other private evidence.
+          required: true

.github/ISSUE_TEMPLATE/config.yml

@@ -0,0 +1,2 @@
+blank_issues_enabled: false
+contact_links: []

.github/ISSUE_TEMPLATE/feature_request.yml

@@ -0,0 +1,34 @@
+name: Feature request
+description: Propose an evidence-backed improvement to the engine or its adapters.
+title: "[Feature]: "
+labels:
+  - enhancement
+body:
+  - type: textarea
+    id: problem
+    attributes:
+      label: What problem should this solve?
+      description: Describe the repeated workflow friction or missing capability.
+    validations:
+      required: true
+  - type: textarea
+    id: proposal
+    attributes:
+      label: Proposed change
+      description: Explain the behavior you would like and where it belongs.
+    validations:
+      required: true
+  - type: textarea
+    id: alternatives
+    attributes:
+      label: Alternatives considered
+      description: Note any existing command, adapter, or workaround you tried.
+  - type: checkboxes
+    id: boundaries
+    attributes:
+      label: Design boundaries
+      options:
+        - label: This preserves local-first evidence handling and explicit human review.
+          required: true
+        - label: This can remain provider-neutral, or the provider-specific behavior is isolated to an adapter.
+          required: true

.github/PULL_REQUEST_TEMPLATE.md

@@ -0,0 +1,18 @@
+## What changed
+
+<!-- Describe the change and the user-visible outcome. -->
+
+## Why
+
+<!-- Link the issue or summarize the evidence that motivated this change. -->
+
+## Validation
+
+<!-- List the exact commands or checks you ran. -->
+
+- [ ] Tests pass.
+- [ ] The bundled Agent Skill validates.
+- [ ] Documentation reflects any user-facing behavior change.
+- [ ] No secrets, raw conversation history, or private evidence are included.
+- [ ] Provider-specific behavior stays in an adapter; the core remains provider-neutral.
+- [ ] Skill changes remain proposal-based and require explicit human review.