Dotta
92c2cdaa5b
docs(evals): link published historical and remote setup reports
2026-10-27 11:48:17 -05:00
Dotta
0dd8e41800
Bind new remote intents to user-authorized AWS fleet fallback
2026-10-27 10:24:23 -05:00
Dotta
3084702f49
test(evals): allow authorized AWS fleet allocation fallback
2026-10-27 10:23:57 -05:00
Dotta
9414380536
test(evals): retain closed remote Docker qualification diagnostics
2026-10-27 10:00:17 -05:00
Dotta
28bdd1ee39
Retain Spot database qualification refusal as separate operation
2026-10-27 09:59:28 -05:00
Dotta
4a0d4252dd
Record remote host refusal without consuming case initials
2026-10-27 09:30:39 -05:00
Dotta
4e93c384ed
Validate remote chat cell provenance before campaign import
2026-10-27 09:19:17 -05:00
Dotta
6175603fd3
fix(evals): validate exact remote PostgreSQL artifact tuples
2026-10-27 09:10:02 -05:00
Dotta
e439748839
feat(evals): gate remote chat on verified Spot and amd64 PostgreSQL
2026-10-27 09:06:20 -05:00
Dotta
279af36535
Close historical local campaign and record remote execution steering
2026-10-27 09:04:27 -05:00
Dotta
9079e0f20d
evals: select OpenCode DeepSeek for remote chat catalog
2026-10-27 08:55:33 -05:00
Dotta
c911d97f4a
fix(evals): deliver selected status card context and current blocker facts
2026-10-27 08:44:44 -05:00
Dotta
23a561cbd8
fix(evals): accept grounded stale-card wording paraphrases
2026-10-27 08:39:41 -05:00
Dotta
e29c9cce65
fix(evals): scaffold preview plugins through the owned source CLI
2026-10-27 08:35:37 -05:00
Dotta
23ca3f56a5
Record 146 processed initials and qualified private-input handoff
2026-10-27 08:31:44 -05:00
Dotta
9660347073
fix(evals): follow scoped runtime session and retry controls
2026-10-27 08:30:57 -05:00
Dotta and Paperclip
fa25178f36
test(evals): accept genuine descriptive pending secret proposal cards
...
Do not misclassify absent duplicate prose as a false completion claim. Require the actual visible human-only source-run-bound proposal card, exact secret/consumer/reference identity, and independent absent binding before either human decision. Contradictory prose, missing authority or wrong-card evidence still fail.
Co-Authored-By: Paperclip <noreply@paperclip.ing >
2026-10-27 08:27:38 -05:00
Dotta
0ae0fc89ba
fix(evals): qualify routine prerequisites and selected webhook context
2026-10-27 08:21:10 -05:00
Dotta
2880e3e6fc
test(evals): minimize repository observation metadata
2026-10-27 08:18:09 -05:00
Dotta and Paperclip
6c09422364
test(evals): review governed tools in actual composer takeover
...
Open the exact persisted pending interaction and verify the selected synthetic app/action in the production takeover section. Scope all decisions to that review instead of waiting for a nonexistent dialog; preserve actual approval and provider-call oracles.
Co-Authored-By: Paperclip <noreply@paperclip.ing >
2026-10-27 08:13:17 -05:00
Dotta
ccb6133d64
fix(evals): bind post-stop readiness to the actual source workspace
2026-10-27 08:07:17 -05:00
Dotta
74558fb8c0
feat(evals): project conservative retained campaign stage proofs
2026-10-27 08:01:40 -05:00
Dotta
96b4ec91cf
test(evals): support qualified owned Docker instance operations
2026-10-27 07:54:19 -05:00
Dotta and Paperclip
420e7fcc6e
test(evals): scope boundary manifest fingerprints to frozen scenario
...
Bind the selected canonical scenario projection while excluding mutable campaign attempt and usage metadata. Preserve actual executable dependencies, conservative unknown-source closure, descriptor and boundary target identity.
Co-Authored-By: Paperclip <noreply@paperclip.ing >
2026-10-27 07:52:12 -05:00
Dotta
3cdf48f17a
test(evals): preserve 124 processed initial checkpoint
2026-10-27 07:49:25 -05:00
Dotta
643447b54d
fix(evals): consume search probe response bodies once
2026-10-27 07:44:09 -05:00
Dotta
8565913082
fix(evals): supply the selected historical developer lab link
2026-10-27 07:41:42 -05:00
Dotta
e74b11e800
fix(evals): preserve historical import data and denied inbox paths
2026-10-27 07:39:36 -05:00
Dotta and Paperclip
699c9a4b4e
test(evals): observe public scoped gateway rotation handoffs
...
Co-Authored-By: Paperclip <noreply@paperclip.ing >
2026-10-27 07:32:20 -05:00
Dotta
9bc325b481
test(evals): retain 118 processed initial checkpoint
2026-10-27 07:29:52 -05:00
Dotta and Paperclip
ab6e73f078
test(evals): deliver selected secure account as factual human context
...
Co-Authored-By: Paperclip <noreply@paperclip.ing >
2026-10-27 07:24:02 -05:00
Dotta and Paperclip
a6dd3301b8
test(evals): decline the actual addressed Notion card
...
Co-Authored-By: Paperclip <noreply@paperclip.ing >
2026-10-27 07:21:21 -05:00
Dotta
78d113f3a7
fix(evals): preserve supported CLI task status and owned feedback gate
2026-10-27 07:17:18 -05:00
Dotta and Paperclip
70e39f41f3
test(evals): isolate source funding from Morgan budget incidents
...
Prepare only company and source-agent public funding before incident prehistory. Independently preserve Morgan policy, incident and pause context and verify the canonical source has no extraneous funding gate.
Co-Authored-By: Paperclip <noreply@paperclip.ing >
2026-10-27 07:16:50 -05:00
Dotta and Paperclip
fb23a1ee76
test(evals): qualify budget prerequisites and company package upload
...
Preserve the original failures while distinguishing nonretryable native billing debt and stopping before a human action can mask unauthorized budget changes. Admit only the exact declared company-preview human upload with existing independent receipt checks.
Co-Authored-By: Paperclip <noreply@paperclip.ing >
2026-10-27 07:15:41 -05:00
Dotta
aa86725867
docs(evals): record 110 processed initial checkpoint
2026-10-27 07:15:16 -05:00
Dotta
00a3a05b22
fix(evals): deliver factual antecedents through human query attachments
2026-10-27 07:08:05 -05:00
Dotta
2c907c0941
fix(evals): qualify case permissions and cancelled-run accounting
2026-10-27 07:07:22 -05:00
Dotta and Paperclip
3032aa27d1
test(evals): follow precise named board test handoffs
...
Co-Authored-By: Paperclip <noreply@paperclip.ing >
2026-10-27 06:57:10 -05:00
Dotta
ff31d399e8
Correct task-bound runtime gates and reviewed boundary evidence
2026-10-27 06:54:11 -05:00
Dotta and Paperclip
beca0672fb
test(evals): isolate private device authorization browser steps
...
Co-Authored-By: Paperclip <noreply@paperclip.ing >
2026-10-27 06:51:00 -05:00
Dotta
30f854b0bc
test(evals): retain twelve independent initial outcomes
2026-10-27 06:49:37 -05:00
Dotta
5f92b5a730
docs(evals): preserve device privacy stop checkpoint
2026-10-27 06:48:36 -05:00
Dotta and Paperclip
7b1ce36f33
test: fingerprint spawned supervisor and source entrypoints
...
Co-Authored-By: Paperclip <noreply@paperclip.ing >
2026-10-27 06:45:37 -05:00
Dotta and Paperclip
06ddb87a9a
test: quarantine invitation enrollment before unsafe token-path acceptance
...
Co-Authored-By: Paperclip <noreply@paperclip.ing >
2026-10-27 06:42:29 -05:00
Dotta
155cf4ce15
fix(evals): separate commentary reader and run evidence owner
2026-10-27 06:33:57 -05:00
Dotta
413de52360
Record authorized independent initial continuation and strict invitation quarantine
2026-10-27 06:19:01 -05:00
Dotta
7a2fc8d200
Preserve four initial outcomes and invitation privacy stop checkpoint
2026-10-27 06:14:47 -05:00
Dotta and Paperclip
33e9a672ce
test: require an authorized cross-company chat handoff
...
Co-Authored-By: Paperclip <noreply@paperclip.ing >
2026-10-27 06:06:43 -05:00
Dotta
bf212430d1
test(e2e): require complete observed Notion choice metadata
2026-10-27 06:04:17 -05:00
Dotta