Commit Graph
324 Commits
Author SHA1 Message Date
Dotta 05b29a381c test(evals): bind one genuine Slack distractor setup exchange 2026-10-26 23:35:15 -05:00
Dotta 9d850991d9 test(evals): prepare owned real Slack endpoint fixtures 2026-10-26 23:32:28 -05:00
Dotta e4b62f48aa test(evals): author remaining internal catalog journeys 2026-10-26 23:30:48 -05:00
Dotta e670ad77ab test(evals): bind scoped Slack identity and audited AgentMail recipient 2026-10-26 23:20:40 -05:00
Dotta 85f49cb351 test(evals): add signed disposable Slack relay for genuine replay proof 2026-10-26 23:11:36 -05:00
Dotta 83ed9d1477 test(evals): require actual unsupported Slack file phase after transfer 2026-10-26 23:10:33 -05:00
Dotta 9c1e30e2c9 test(evals): bind canonical Slack ingress to actual native source 2026-10-26 23:06:56 -05:00
Dotta c2b8cb697a test(evals): qualify finite Slack companion provider and source contracts 2026-10-26 23:04:03 -05:00
Dotta 7d357062f5 test(evals): probe unsupported effective skill-use prerequisites 2026-10-26 23:01:42 -05:00
Dotta fccad2fb5b test(evals): exercise real card refresh failure and retained output 2026-10-26 22:58:50 -05:00
Dotta 1fc2c961a2 test(evals): include named profile prerequisite in inventory controls 2026-10-26 22:56:20 -05:00
Dotta f2c3d07357 test(evals): author named Notion read-only profile wizard 2026-10-26 22:55:18 -05:00
Dotta 49dbaf516b test: cover card explanation after real source removal 2026-10-26 22:52:51 -05:00
Dotta c27280d1f0 test(evals): author secure company secret and Morgan binding journey 2026-10-26 22:52:12 -05:00
Dotta 1ff71fdb8b test: cover authenticated routine webhook with isolated human setup 2026-10-26 22:49:04 -05:00
Dotta 994897f542 test: retain selected input blockers and authenticate private oracle identity 2026-10-26 22:48:22 -05:00
Dotta 8c9b4ac273 test(evals): author actual Notion catalog choice journey 2026-10-26 22:46:19 -05:00
Dotta 509fcf3514 test(evals): reject contradictory fresh preview success claims 2026-10-26 22:44:00 -05:00
Dotta 25914805ac test(evals): author real plugin preview tool journey 2026-10-26 22:41:06 -05:00
Dotta 7733c54e21 test(evals): verify rendered native artifacts without post-query direction 2026-10-26 22:36:03 -05:00
Dotta 01e7e1b38a test: bind scoped MCP cancellation source revision 2026-10-26 22:35:17 -05:00
Dotta b22ad6fc9c test(evals): author supported native interruption with accepted fresh direction 2026-10-26 22:30:02 -05:00
Dotta 5812962351 test: cover genuine tracked accessibility skill revision update 2026-10-26 22:27:40 -05:00
Dotta 73c414d968 test: accept natural case closure and refusal phrasing 2026-10-26 22:27:40 -05:00
Dotta 1b90c8d7c0 test(evals): preserve immutable responses across ordinary-only queue actions 2026-10-26 22:24:42 -05:00
Dotta 3cc814bdb4 test(evals): add real scoped external assistant model companion 2026-10-26 22:24:05 -05:00
Dotta 1d62a20222 test: reject contradictory case closure claims by clause 2026-10-26 22:22:18 -05:00
Dotta fcb04b4d0f test(evals): author immutable answer delivery beside ordinary queue controls 2026-10-26 22:20:45 -05:00
Dotta 3350d46798 test: cover genuine terminal case closure and review gate 2026-10-26 22:20:38 -05:00
Dotta cf4aad60e0 test(evals): author exhausted retry with real admission gate phases 2026-10-26 22:13:45 -05:00
Dotta 01df721473 test: add finite scoped external assistant catalog surface 2026-10-26 22:13:26 -05:00
Dotta 0a97b24b64 test(evals): author genuine owned native infrastructure recovery fixtures 2026-10-26 22:09:30 -05:00
Dotta 02c0b64b72 test(evals): disconnect exact Notion account through public UI 2026-10-26 22:08:54 -05:00
Dotta 5e8eefb199 test: preserve ordinary case handoff eligibility in facts fixture 2026-10-26 22:07:59 -05:00
Dotta 3232b9d129 test: cover attributed case diagnosis history 2026-10-26 22:04:53 -05:00
Dotta 80a580212c test(evals): verify Morgan-only Notion audience through actual picker 2026-10-26 22:00:14 -05:00
Dotta fdd772c7a7 test: cover corrected case facts and concurrent reread 2026-10-26 21:58:21 -05:00
Dotta 89163d7f73 test(evals): author governed independent native review journey 2026-10-26 21:57:06 -05:00
Dotta 6d04ec8d5f test(evals): rotate gateway tokens through isolated real UI flow 2026-10-26 21:56:37 -05:00
Dotta bcd7ca5e77 test(evals): distinguish substantive blue steering and observed branch 2026-10-26 21:52:31 -05:00
Dotta ac8ac79d68 test(evals): author actual native same-turn steering journey 2026-10-26 21:48:43 -05:00
Dotta 37b72e28e3 test(evals): provision genuine plugin-managed routine and preserve its controls 2026-10-26 21:48:20 -05:00
Dotta edbf426db3 test(evals): distinguish unsupported skill action from unfinished denial coverage 2026-10-26 21:48:20 -05:00
Dotta 8a5da0ab01 test(evals): exercise actual board test-as-agent and provider outcome 2026-10-26 21:47:45 -05:00
Dotta f097a9da02 test(evals): accept independently accessible delegation briefs 2026-10-26 21:45:07 -05:00
Dotta d4f02a1e67 test(evals): reject premature and wrong-client rotation guidance 2026-10-26 21:44:36 -05:00
Dotta c211c554b4 test(evals): verify scoped token rotation and Notion assignment continuation 2026-10-26 21:41:11 -05:00
Dotta fe05c54c11 test(evals): probe real skill assignment authority and future version delivery 2026-10-26 21:40:00 -05:00
Dotta fa7eba012e evals: carry real launch brief through one Morgan handoff and result 2026-10-26 21:36:20 -05:00
Dotta 29485d6a45 test(evals): verify real team installation and member configuration 2026-10-26 21:34:13 -05:00