@@ -0,0 +1,8 @@
+blank_issues_enabled: false
+contact_links:
+  - name: Security vulnerability (email)
+    url: mailto:zxhwolfe@gmail.com?subject=Topic%20Intelligence%20security%20report
+    about: Report vulnerabilities privately. Do not open a public Issue.
+  - name: Privacy concern (email)
+    url: mailto:zxhwolfe@gmail.com?subject=Topic%20Intelligence%20privacy%20concern
+    about: Send privacy concerns privately without credentials or unnecessary sensitive data.
@@ -0,0 +1,44 @@
+name: Feature request
+description: Suggest a behavior grounded in a real creator, research, or editorial workflow.
+title: "[Feature]: "
+labels: ["enhancement"]
+body:
+  - type: markdown
+    attributes:
+      value: |
+        Describe the real workflow without including credentials, complete conversations, private source material, or customer data.
+  - type: textarea
+    id: scenario
+    attributes:
+      label: Real use case
+      description: Who is doing what, in which Host, and what outcome do they need?
+    validations:
+      required: true
+  - type: textarea
+    id: blocker
+    attributes:
+      label: Current obstacle
+      description: What cannot be completed with the current three modes?
+    validations:
+      required: true
+  - type: textarea
+    id: behavior
+    attributes:
+      label: Suggested behavior
+      description: What should Topic Intelligence do differently?
+    validations:
+      required: true
+  - type: textarea
+    id: privacy
+    attributes:
+      label: Privacy and telemetry impact
+      description: Would this require content, conversation, account, or usage data? Prefer privacy-preserving designs.
+    validations:
+      required: true
+  - type: checkboxes
+    id: safety
+    attributes:
+      label: Public-data confirmation
+      options:
+        - label: I removed secrets, personal data, client data, and complete conversation content.
+          required: true
@@ -0,0 +1,55 @@
+name: Installation failure
+description: Report a problem installing or discovering Topic Intelligence in a supported host.
+title: "[Install]: "
+labels: ["installation"]
+body:
+  - type: markdown
+    attributes:
+      value: |
+        Do not paste API keys, passwords, tokens, complete conversations, customer data, or other sensitive information. Redact errors before submitting.
+  - type: input
+    id: version
+    attributes:
+      label: Topic Intelligence version
+      description: Use the Release version and ZIP filename.
+      placeholder: v0.3.0 / topic-intelligence-0.3.0.zip
+    validations:
+      required: true
+  - type: dropdown
+    id: host
+    attributes:
+      label: Host
+      options:
+        - ChatGPT Web
+        - Codex
+        - Another compatible Agent Skills host
+    validations:
+      required: true
+  - type: textarea
+    id: environment
+    attributes:
+      label: Environment
+      description: Plan/workspace type if relevant, operating system, browser or host version. Do not include account identifiers.
+    validations:
+      required: true
+  - type: textarea
+    id: steps
+    attributes:
+      label: Installation steps
+      description: List the steps you took and where the installation stopped.
+    validations:
+      required: true
+  - type: textarea
+    id: error
+    attributes:
+      label: Redacted error
+      description: Paste only the minimum redacted error text. Do not include a complete conversation or credentials.
+    validations:
+      required: true
+  - type: checkboxes
+    id: safety
+    attributes:
+      label: Public-data confirmation
+      options:
+        - label: I removed secrets, personal data, client data, and complete conversation content.
+          required: true
@@ -0,0 +1,54 @@
+name: Result quality
+description: Report a selection or brief result that did not meet the documented evidence boundary.
+title: "[Quality]: "
+labels: ["result-quality"]
+body:
+  - type: markdown
+    attributes:
+      value: |
+        Summarize the problem. Do not publish complete conversations, private source material, credentials, or customer data.
+  - type: dropdown
+    id: mode
+    attributes:
+      label: Mode
+      options:
+        - Selection only
+        - Supplied current topic to brief
+        - Selection followed by brief
+    validations:
+      required: true
+  - type: input
+    id: host_version
+    attributes:
+      label: Host and Skill version
+      placeholder: ChatGPT Web / v0.3.0
+    validations:
+      required: true
+  - type: textarea
+    id: expected
+    attributes:
+      label: Expected result
+      description: Describe the output shape or behavior you expected.
+    validations:
+      required: true
+  - type: textarea
+    id: actual
+    attributes:
+      label: Actual quality problem
+      description: Provide a short, redacted summary rather than a full transcript.
+    validations:
+      required: true
+  - type: textarea
+    id: evidence_boundary
+    attributes:
+      label: Evidence-boundary concern
+      description: Did the result hide stale/partial data, mix Radar facts with editorial judgment, predict virality, change the topic, or miss must_verify/avoid_claims?
+    validations:
+      required: true
+  - type: checkboxes
+    id: safety
+    attributes:
+      label: Public-data confirmation
+      options:
+        - label: I removed secrets, personal data, client data, and complete conversation content.
+          required: true
@@ -19,7 +19,9 @@ selection-only, supplied-topic brief, and selection-followed-by-brief requests.
 - Rejected duplicate successful Radar requests and required selection workflows
   to use exactly one feed, so compact-scan claims remain trace-verifiable.
 
-The v0.2.0 line remains the last version with manual ChatGPT Web ZIP upload evidence. v0.2.1 and v0.2.2 Host Eval records do not claim ChatGPT Web UI validation.
+### Post-release validation
+
+On 2026-08-12, the final published v0.3.0 ZIP completed a post-release ChatGPT Web smoke with natural user prompts. Selection only, supplied-topic brief, and selection followed by brief all passed at the user-visible level. This is a documentation/evidence update only; the v0.3.0 ZIP contents did not change. Internal command counts and raw traces remain covered by the separate Codex Host Eval and release-evidence gates.
 
 ## [0.2.2] - 2026-08-11
 
@@ -78,7 +80,7 @@ No unreleased changes.
 
 ### Distribution
 
-- `v0.2.2` is the current package line; `v0.2.0` remains immutable and retains the last manual ChatGPT web-upload evidence, while v0.2.1 and v0.2.2 Host Eval results do not claim ChatGPT Web UI validation.
+- At that release stage, `v0.2.2` was the current package line and immutable v0.2.0 retained the latest recorded manual ChatGPT web-upload evidence; v0.2.1 and v0.2.2 Host Eval results did not claim ChatGPT Web UI validation. The final v0.3.0 ZIP later completed its own post-release three-mode ChatGPT Web smoke on 2026-08-12.
 - No Hosted MCP transport is required for public Radar connectivity because standalone ChatGPT Skills already reach the live public read endpoints.
 - A future authenticated App/Plugin/OAuth transport may be introduced only for Premium account-bound capabilities such as server Topic Insight.