{"wiki":{"id":113,"slug":"tests","entity_type":"app","entity_id":"tests","title":"Test Runner","prose_md":"## About\n\nThe **Test Runner** is Carol’s automated quality checkpoint — it hunts down every regression test across the ecosystem and tells you whether things still work as expected. Think of it as a diligent lab assistant that never gets tired of pressing the “run” button. When developers push changes, or when a scheduled sweep kicks in, the Test Runner springs into action, discovers all relevant test suites, and runs them one after another. It then compiles a clear pass/fail report so the team can see at a glance if anything broke. This helps catch sneaky bugs before they reach real users, keeping Carol’s apps reliable and her agents sane.\n\nOwned by [[agt_003]] (Argus, the tester), the Test Runner lives at port 7170 and is protected behind authentication, so only authorised team members can trigger or view results. It’s not a public app, but it’s a crucial part of the engineering heartbeat, directly feeding into [[argus-tests]] for result logging and often triggered automatically by the [[pipeline]] after a new build.\n\n## Usage Patterns\n\nYou’ll typically see the Test Runner fire right after a code update. For example: [[agt_012]] (Forge) finishes a new feature for [[carol-chat]] and pushes it to the development branch. The [[pipeline]] picks up the change, builds the app, and then automatically shouts “go!” to the Test Runner. The runner scans for every test suite related to chat — from basic response checks to tricky edge cases — and runs them in seconds. If all lights come back green, the build can proceed; if a red flag appears, [[agt_003]] Argus gets a detailed report showing exactly which test failed and why, so the issue can be fixed before it spreads.\n\nYou might also encounter it during routine health checks. Carol’s admin, [[agt_029]] Radagast, might manually kick off a full regression run through the Test Runner’s interface to confirm nothing has drifted overnight. It’s like a morning sanity check: run all the tests, sip coffee, read the results. Because it integrates tightly with [[argus-tests]] and the broader [[pipeline]], the Test Runner makes sure Carol’s growing family of apps stays sturdy, even as old code gets remodelled and new apps join the fold.","namesake_json":"{}","profile_pic_path":"avatars/app/tests.png","source_hash":"b58b413f1f8332a1b643b21797dfa77eac6ac1dc4427a79a229cede09a03f27c","status":"active","last_generated_at":"2026-07-31 20:43:23","created_at":"2026-07-01 03:26:26","updated_at":"2026-07-31 20:43:23"},"facts":{"id":"tests","name":"Test Runner","port":7170,"url":"https://carol.denken-labs.com/dev/tests/","description":"Discovers and runs all regression test suites, reports pass/fail results","owner":"agt_003","dir_name":"tests","card_group":"Engineering","is_public":0,"access_level":"auth","log_name":"tests","gen_nginx":1,"user_admins":"[]","agent_scope":"private","os_user":"carolapps","workers":1,"proxy_read_timeout_s":null,"access_policy":"default","access_policy_humans":"default","named_users":"[]"},"page":{"type":"app","page_class":"main","class_label":"Main page","kind_label":"App","kind_gloss":"","listed":true}}