64dfad6d53
feat(runs): add typed results and model aliases (#2043)
samandCopilot2026-11-18 23:47:35 +11:00
6a3bf5a71a
refactor(governance): open role identifiers and guard policy lockout with Cedar (#2017)
v1.63.0
sam2026-11-18 02:29:06 +11:00
0218e4ec89
feat(governance): move identity, org and Cedar access engine into governance (#2017)
sam2026-11-18 02:09:29 +11:00
ce70d4f3ee
feat(runs): workflow documents take today's step shape, references and when (#2039)
sam2026-11-17 18:43:45 +11:00
83613d902a
docs: format the docs
sam2026-11-17 17:53:58 +11:00
1026d5e3b6
feat(httpapi): the API is described by an OpenAPI document generated with Huma (#2028)
sam2026-11-16 20:05:05 +11:00
ee4c3f9c16
docs(agents): old code moves into its module; a rewrite or a deletion needs a verdict in the move map (#1977)
sam2026-11-16 15:52:21 +11:00
21de660ae1
test(system): the rotation fixture uses a listen address without an IP (#2026)
sam2026-11-16 13:48:07 +11:00
52b027955b
feat(connectors): a secret setting is resolved through its engine before the extension starts (#2026)
sam2026-11-16 13:46:48 +11:00
e2917420fb
feat(ui): the dashboard edits every control-plane setting (#2026)
sam2026-11-16 13:45:20 +11:00
c226c4f686
feat(system): bootstrap values are read-only and the admin token rotates into the config (#2026)
sam2026-11-16 13:45:18 +11:00
9e2cf252ac
feat(judgement): the Controls API reports each guard's judge, its standing and its eval date (#2026)
sam2026-11-16 13:45:18 +11:00
9fe446a426
feat(governance): the agent model names a credential and answers a test call (#2026)
sam2026-11-16 13:45:15 +11:00
20d81258ac
feat(events): sources, triggers and filtered events are listed for the dashboard (#2026)
sam2026-11-16 13:45:15 +11:00
053ed58c08
feat(connectors): an extension declares the settings it reads, and each accepted package is listed (#2026)
sam2026-11-16 13:45:12 +11:00
546e5cd6ef
refactor(ui): hard rules and change-request rules are Controls, not Policies, which stays the access model (archie #1977)
sam2026-11-16 13:23:09 +11:00
124d2033fc
feat(guard): a judge record keeps the request, response, path, status and latency of every model call
sam2026-11-16 12:03:23 +11:00
46372b51d5
feat(runs,ui): the trail shows each model request and response, connector calls and delivery attempts
sam2026-11-16 12:04:41 +11:00
db36d853ce
feat(deliveries): every attempt keeps its response, its error and how long it took
sam2026-11-16 12:04:39 +11:00
4b97ac5841
feat(audit): a connector call keeps the request core sent and the response it received
sam2026-11-16 12:04:36 +11:00
e909aee155
refactor(ui): the Unmatched page lives at /unmatched, matching the nav label (#1994)
sam2026-11-16 12:06:04 +11:00
564339e1aa
feat(events,runs): an event keeps the payload it arrived in, and every model call records the request it sent via ai-sdk v0.8.1 (audit requirement)
sam2026-11-16 11:54:50 +11:00
e45fee9bc9
fix(runs): a held action is described to the agent as waiting for approval, not a person's
sam2026-11-16 11:48:05 +11:00
4cdb5996c5
feat(guard): a model judge's answer, basis and eval standing are recorded in every verdict, clean answers included (audit requirement)
sam2026-11-16 11:12:15 +11:00
4657ff2646
fix(runs): a workflow summary names approval, not a person, as what holds an action (#2008)
sam2026-11-16 10:09:59 +11:00
703d5376bb
feat(ui): the Unmatched page shows each group's draft with its summary, dry-run trail and publish (#2008)
sam2026-11-16 10:04:44 +11:00
dd80500e7a
feat(drafts): a draft bundles trigger, workflow and policy with a summary and a dry run, and publishes them (#2008)
sam2026-11-16 10:04:44 +11:00
1e21aa975d
refactor(runs): a workflow is a stored document of agent, action, wait-for-approval, notify and sleep steps, and investigate ships as one; dry mode judges without executing (#2008)
sam2026-11-16 10:04:44 +11:00
66b63260e9
feat(events): a group's stored events can be replayed, and a trigger saved, through the module's API (#2008)
sam2026-11-16 10:04:44 +11:00
470015483e
feat(judgement): a dry decision is recorded apart, counts toward nothing and never executes a changing action, and a draft's policy can overlay the set in force (#2008)
sam2026-11-16 10:04:44 +11:00
42f0688f84
feat(pg): tables for stored workflow documents, dry runs and drafts (#2008)
sam2026-11-16 10:04:44 +11:00
b6f7059212
feat(storepkg,ui): a package can declare a multi-line text setting (#2006)
sam2026-11-16 09:56:10 +11:00
88afd1ebc9
feat(guard,runs): a guard is shown the agent's diagnosis and the evidence it read before proposing an action (#2006)
sam2026-11-16 09:55:14 +11:00
a4a4e97f07
feat(judgement): a target's owner is looked up from the connectors at decision time, and an unknown owner is a finding that sends the request to the default owner (#1991)
sam2026-11-16 09:49:33 +11:00
4bc0d54341
refactor(governance): check escalation deadlines in their own function, keeping validate under the complexity limit (#1985)
sam2026-11-16 09:44:30 +11:00
dabe0b4f7c
refactor(approvals): judgement's change-request rules route an approval, and without a connector it goes to the channels (#1985)
sam2026-11-16 09:41:39 +11:00
28733c8522
feat(approvals,runs): an unanswered approval escalates step by step and then expires, a discarded approval never counts, and a change-request seam routes and withdraws (#1985)
sam2026-11-16 09:35:39 +11:00
4b1ae4babf
feat(governance): each escalation step has its own deadline, and the path skips anyone below the authority an action needs (#1985)
sam2026-11-16 09:35:39 +11:00
50b75f6987
feat(events): unmatched events are grouped by source, type and key fields, and Archie's own deliveries never group (#1994)
sam2026-11-16 09:32:24 +11:00
0c8b673615
refactor(archie): build the policy sets in their own constructor so Run stays within its statement limit
sam2026-11-16 09:40:26 +11:00
ef62dcc97a
feat(judgement): impact, provenance, timing and cost guards from data, with business hours, blackouts and action budgets as settings (#1988)
sam2026-11-16 09:30:47 +11:00
82cfc1de5b
feat(judgement): hard rules and change-request rules as versioned Cedar policy sets (#1989)
sam2026-11-16 09:30:39 +11:00
983f462740
fix(governance): the settings and people responses no longer fail to encode on an unsupported json tag
sam2026-11-16 05:23:38 +11:00
46edda9961
refactor(runs): one conclude step ends an investigation, keeping Investigate under the complexity limit
sam2026-11-16 05:21:48 +11:00