chore: prepare Strandly bug verification pilot

.agents/learnings/README.md

@@ -0,0 +1,33 @@
+# Agent Learnings
+
+This directory contains durable, repository-specific constraints learned from completed agent runs. It is an evidence
+log, not a scratchpad or a replacement for `AGENTS.md`.
+
+Agents should consult relevant entries before changing or reviewing the affected area. Rules that apply broadly and
+remain stable should be promoted to `AGENTS.md` through a normal pull request.
+
+## Admission Criteria
+
+A learning must:
+
+- Link to the source run and the issue or pull request that supplied evidence.
+- Describe a repeatable repository constraint, not a one-off implementation.
+- Explain how the evidence changes a future implementation or review decision.
+- Avoid credentials, customer data, internal-only links, and copied user data.
+- Be submitted and reviewed through the normal pull request process.
+
+Do not record style preferences, speculative advice, or conclusions based only on an agent's own output.
+
+## Format
+
+```markdown
+# Actionable rule
+
+- **Source run:** <run identifier>
+- **Source issue or PR:** <public GitHub link>
+- **Affected area:** <paths, subsystem, or workflow>
+
+State the durable constraint and the action a future agent should take.
+
+Explain the evidence, the previous failure mode, and why the rule generalizes.
+```

.github/ISSUE_TEMPLATE/bug_report.yml

@@ -1,73 +1,152 @@
 name: Bug Report
-description: Report a bug or unexpected behavior
+description: Report reproducible unexpected behavior in the AgentCore CLI
+title: '[BUG] '
 labels: ['bug']
 body:
   - type: markdown
     attributes:
       value: |
-        Thanks for taking the time to report a bug. Please fill out the sections below so we can reproduce and fix the issue.
+        Include a minimal reproduction and exact environment details. Remove
+        credentials, tokens, account IDs, customer data, and other sensitive
+        information from commands, configuration, and logs.
+
+  - type: checkboxes
+    id: checks
+    attributes:
+      label: Before submitting
+      options:
+        - label: I searched existing issues and did not find a duplicate.
+          required: true
+        - label: I tested with the latest available AgentCore CLI release.
+          required: true
+        - label: I removed credentials and sensitive data from this report.
+          required: true
 
   - type: textarea
     id: description
     attributes:
       label: Description
-      description: A clear and concise description of the bug.
-      placeholder: Describe what happened...
+      description: Describe the incorrect behavior and its impact.
+      placeholder: What happened, and what does it prevent you from doing?
+    validations:
+      required: true
+
+  - type: input
+    id: cli-version
+    attributes:
+      label: CLI version
+      description: Output of `agentcore --version`.
+      placeholder: '1.2.3'
+    validations:
+      required: true
+
+  - type: input
+    id: node-version
+    attributes:
+      label: Node.js version
+      description: Output of `node --version`.
+      placeholder: 'v22.17.0'
+    validations:
+      required: true
+
+  - type: dropdown
+    id: installation-method
+    attributes:
+      label: Installation method
+      options:
+        - npm global install
+        - npx
+        - Built from source
+        - Package tarball
+        - Other
+    validations:
+      required: true
+
+  - type: input
+    id: operating-system
+    attributes:
+      label: Operating system
+      description: Include the operating system name, version, and architecture.
+      placeholder: 'macOS 15.5 arm64'
+    validations:
+      required: true
+
+  - type: input
+    id: aws-region
+    attributes:
+      label: AWS Region
+      description: Region used when the problem occurred, or `N/A` for local-only behavior.
+      placeholder: 'us-west-2'
+    validations:
+      required: true
+
+  - type: dropdown
+    id: resource-dependency
+    attributes:
+      label: Resource dependency
+      description: Does the reproduction require live AWS resources?
+      options:
+        - No, it reproduces locally
+        - Yes, it requires live AWS resources
+        - Both local and live AWS behavior are affected
+        - Unknown
     validations:
       required: true
 
   - type: textarea
     id: steps
     attributes:
-      label: Steps to Reproduce
-      description: Steps to reproduce the behavior.
+      label: Minimal reproduction
+      description: |
+        Provide the smallest complete sequence that reproduces the problem,
+        including project setup and commands. Use placeholders for identifiers.
       placeholder: |
-        1. Run `agentcore ...`
-        2. Select ...
-        3. Observe ...
+        1. Create a project with `agentcore create ...`
+        2. Add this minimal configuration: ...
+        3. Run `agentcore ...`
+        4. Observe ...
+      render: shell
     validations:
       required: true
 
   - type: textarea
     id: expected
     attributes:
-      label: Expected Behavior
-      description: What you expected to happen.
+      label: Expected behavior
+      description: State the exact result you expected.
     validations:
       required: true
 
   - type: textarea
     id: actual
     attributes:
-      label: Actual Behavior
-      description: What actually happened.
+      label: Actual behavior and error output
+      description: Include the complete relevant error and stack trace after removing sensitive data.
+      render: shell
     validations:
       required: true
 
-  - type: input
-    id: cli-version
-    attributes:
-      label: CLI Version
-      description: Output of `agentcore --version`
-      placeholder: '1.0.0'
-    validations:
-      required: false
-
   - type: dropdown
-    id: os
+    id: frequency
     attributes:
-      label: Operating System
+      label: Reproduction frequency
       options:
-        - macOS
-        - Windows
-        - Linux
+        - Every time
+        - Intermittently
+        - Once
+        - Unknown
     validations:
       required: true
 
+  - type: input
+    id: last-working-version
+    attributes:
+      label: Last known working version
+      description: If this is a regression, provide the last CLI version that worked.
+      placeholder: 'N/A or 1.2.2'
+
   - type: textarea
     id: context
     attributes:
-      label: Additional Context
-      description: Any other context, screenshots, or logs that might help.
-    validations:
-      required: false
+      label: Additional context
+      description: Add sanitized configuration, screenshots, or related issue links.

.github/workflows/auto-close-unverified-bugs.yml

@@ -0,0 +1,139 @@
+name: Auto Close Unverified Bugs
+
+on:
+  schedule:
+    - cron: '0 14 * * 1-5'
+  workflow_dispatch:
+    inputs:
+      dry_run:
+        description: Log actions without changing issues
+        required: false
+        default: true
+        type: boolean
+
+permissions:
+  contents: read
+  issues: write
+
+jobs:
+  close-unverified-bugs:
+    runs-on: ubuntu-latest
+    timeout-minutes: 10
+    steps:
+      - name: Process issues awaiting reporter information
+        uses: actions/github-script@v9
+        env:
+          LABEL_NAME: autoclose in 7 days
+          DAYS_TO_WAIT: '7'
+          MANUAL_DRY_RUN: ${{ inputs.dry_run || false }}
+        with:
+          script: |
+            const labelName = process.env.LABEL_NAME;
+            const waitMs =
+              Number(process.env.DAYS_TO_WAIT) * 24 * 60 * 60 * 1000;
+            const dryRun =
+              context.eventName === "workflow_dispatch" &&
+              process.env.MANUAL_DRY_RUN === "true";
+
+            const candidates = await github.paginate(
+              github.rest.issues.listForRepo,
+              {
+                owner: context.repo.owner,
+                repo: context.repo.repo,
+                state: "open",
+                labels: labelName,
+                per_page: 100,
+              },
+            );
+
+            const issues = candidates.filter((issue) => !issue.pull_request);
+            core.info(`Found ${issues.length} issue(s) with "${labelName}"`);
+
+            for (const issue of issues) {
+              const events = await github.paginate(
+                github.rest.issues.listEvents,
+                {
+                  owner: context.repo.owner,
+                  repo: context.repo.repo,
+                  issue_number: issue.number,
+                  per_page: 100,
+                },
+              );
+
+              const labelEvents = events
+                .filter(
+                  (event) =>
+                    event.event === "labeled" &&
+                    event.label?.name === labelName,
+                )
+                .sort(
+                  (left, right) =>
+                    new Date(right.created_at) - new Date(left.created_at),
+                );
+
+              if (labelEvents.length === 0) {
+                core.warning(
+                  `Skipping #${issue.number}: label event was not found`,
+                );
+                continue;
+              }
+
+              const labeledAt = new Date(labelEvents[0].created_at);
+              const comments = await github.paginate(
+                github.rest.issues.listComments,
+                {
+                  owner: context.repo.owner,
+                  repo: context.repo.repo,
+                  issue_number: issue.number,
+                  since: labeledAt.toISOString(),
+                  per_page: 100,
+                },
+              );
+
+              const reporterResponded = comments.some(
+                (comment) =>
+                  comment.user?.login === issue.user?.login &&
+                  new Date(comment.created_at) > labeledAt,
+              );
+
+              if (reporterResponded) {
+                core.info(
+                  `Reporter responded on #${issue.number}; removing label`,
+                );
+                if (!dryRun) {
+                  await github.rest.issues.removeLabel({
+                    owner: context.repo.owner,
+                    repo: context.repo.repo,
+                    issue_number: issue.number,
+                    name: labelName,
+                  });
+                }
+                continue;
+              }
+
+              if (Date.now() - labeledAt.getTime() < waitMs) {
+                core.info(`#${issue.number} is still within the waiting period`);
+                continue;
+              }
+
+              core.info(`Closing #${issue.number} after the waiting period`);
+              if (!dryRun) {
+                await github.rest.issues.createComment({
+                  owner: context.repo.owner,
+                  repo: context.repo.repo,
+                  issue_number: issue.number,
+                  body:
+                    "This issue was automatically closed because the " +
+                    "requested reproduction details were not provided within " +
+                    "7 days. Add the missing information in a new issue if " +
+                    "the problem persists.",
+                });
+                await github.rest.issues.update({
+                  owner: context.repo.owner,
+                  repo: context.repo.repo,
+                  issue_number: issue.number,
+                  state: "closed",
+                  state_reason: "not_planned",
+                });
+              }
+            }

AGENTS.md

@@ -2,6 +2,12 @@
 
 CLI tool for Amazon Bedrock AgentCore. Manages agent infrastructure lifecycle.
 
+## Learned Constraints
+
+Repository-specific constraints derived from completed agent runs live in `.agents/learnings/`. Consult relevant entries
+before changing or reviewing an affected area. New learnings require evidence and normal pull request review; see
+`.agents/learnings/README.md`.
+
 ## Package Structure
 
 ```