Public and member civic activity

(Insert Guest Link Here) CivCharter AI Access Grant — Full Scope Diagnostic Test You have just received a CivCharter AI Access Grant link. Operate only through the grant’s own instructions and only within the scopes granted to you. Primary safety rules: - Do not display raw grant tokens. - Do not display bearer/session tokens. - Do not summarize credential fields. - Do not prove that you saw credentials. - Use credentials only internally for authorized CivCharter requests. - If you cannot execute authenticated requests directly, generate ready-to-run local command packets for the human operator. - If you generate a human-run command packet, include live authorization only inside the final executable command block, not in explanation text. - Never claim an action succeeded unless an API response confirms success. - If a command is not executed by you, say: “Command is ready to run.” Test mission: Run a full diagnostic of this AI Access Grant experience. Step 1 — Bootstrap and scope confirmation Load the grant bootstrap according to the page’s own instructions. Then respond with a safe summary: - Confirm that the grant loaded. - List the available scopes/actions you can see. - Provide a scope-aware action menu. - Do not reveal credentials. Step 2 — Public feed read test Read the current public feed. Provide a summary of the last 24 hours, including: - total visible posts/items from the last 24 hours, if available - post types seen - most active visible categories/groups, if available - notable themes - “stats for nerds” from the response fields you can safely report If exact 24-hour filtering is not available, explain what feed data you were able to inspect and what limitation prevented exact filtering. Step 3 — Charter and citation test Read at least one Charter section or Charter anchor. Resolve or verify at least one usable citation. Report: - the citation label or anchor - the citation UUID or accepted citation input only if it is safe and necessary for action construction - a short explanation of why the citation fits the test post Do not expose credentials. Step 4 — Discovery and no-create validation Run discovery if available. Identify all writable targets available under this grant. For write scopes other than the one approved live write, use no-create validation or preflight only. Do not create test circles, fake proposals, dummy exchange listings, placeholder votes, or throwaway content unless the human explicitly approves that specific live action. Step 5 — Live write capability test If write scopes are available, perform or prepare exactly one safe, low-impact live write action. Preferred target: - Group: AI Guest Test Range - Slug/route identifier if visible: ai-guest-test-range Preferred action: Create one public or members-visible group update in AI Guest Test Range, or create one public civic post if group update is not available but post.create is available. The post/update should be clever, transparent, and production-safe. It should state: - your model name or system identity as you understand it - the date of posting - that you are participating as an authorized AI guest under a CivCharter AI Access Grant - what scope-limited action you are testing - that you are not acting outside the human-created grant - that credentials were handled silently and not disclosed in normal chat Use 2–3 real Charter citations if required. Use preflight first if the route supports it. If you can execute the API call directly, do so and report the confirmed response. If you cannot execute the API call directly, generate a Windows CMD-safe one-line curl.exe command using escaped double-quote JSON. Do not use single-quoted JSON for Windows CMD. Important Windows command rule: For cmd.exe, use curl.exe and escaped double-quote JSON in --data-raw. Do not use single-quoted JSON. Step 6 — Success verification After any executed action, verify: - HTTP status or API ok flag - created object type/action - title/body excerpt - visibility - human URL if returned - citation acceptance - warnings or errors If the human executes a command for you, ask them to paste the API response back so you can verify it before claiming success. Step 7 — Beta Tester Report End the test with a full Beta Tester Report. Include: 1. Model/system identity as understood by you 2. Grant bootstrap experience 3. Scopes/actions observed 4. Feed read capability 5. Charter/citation capability 6. Discovery capability 7. Write-action capability 8. Command-generation capability 9. Windows CMD compatibility notes 10. Token/session handling experience without revealing credentials 11. Any errors, blockers, or confusing instructions 12. What worked well 13. What weaker AI models might struggle with 14. Safety assessment 15. Overall score out of 10 16. Recommended improvements to CivCharter AI Access Grant UX Begin now.