17 Commits
Author SHA1 Message Date
sam 1026d5e3b6 feat(httpapi): the API is described by an OpenAPI document generated with Huma (#2028)
Every /api operation is registered with Huma behind one admin-token
middleware, served at /api/openapi.json with a docs page at /api/docs,
and committed as api/openapi.yaml, which task check keeps current.
2026-11-16 20:05:05 +11:00
sam 9e2cf252ac feat(judgement): the Controls API reports each guard's judge, its standing and its eval date (#2026) 2026-11-16 13:45:18 +11:00
sam 053ed58c08 feat(connectors): an extension declares the settings it reads, and each accepted package is listed (#2026) 2026-11-16 13:45:12 +11:00
sam 2f802e04a4 Merge branch 'worktree-agent-a1bd90c329b751a72' 2026-11-16 12:06:07 +11:00
sam 124d2033fc feat(guard): a judge record keeps the request, response, path, status and latency of every model call 2026-11-16 12:06:05 +11:00
sam 4b97ac5841 feat(audit): a connector call keeps the request core sent and the response it received
Resolve, target facts, invoke, lookup, deliver and parse record the protobuf wire messages as JSON on the context recorder the owning module supplies. Credentials are redacted and a body over 1 MiB keeps its prefix, full length and SHA-256. Judgement stores the resolve and target facts in the decision body and appends the executed invoke to it.
2026-11-16 12:06:05 +11:00
sam 4cdb5996c5 feat(guard): a model judge's answer, basis and eval standing are recorded in every verdict, clean answers included (audit requirement) 2026-11-16 11:12:15 +11:00
sam 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) 2026-11-16 10:05:17 +11:00
sam 88afd1ebc9 feat(guard,runs): a guard is shown the agent's diagnosis and the evidence it read before proposing an action (#2006) 2026-11-16 09:55:14 +11:00
sam 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) 2026-11-16 09:49:33 +11:00
sam ef62dcc97a feat(judgement): impact, provenance, timing and cost guards from data, with business hours, blackouts and action budgets as settings (#1988) 2026-11-16 09:38:34 +11:00
sam 82cfc1de5b feat(judgement): hard rules and change-request rules as versioned Cedar policy sets (#1989) 2026-11-16 09:38:07 +11:00
sam 55077620bb feat(judgement,connectors): guard extension contract; every running guard judges with the built-ins and records its name and digest (archie #1986) 2026-11-16 05:17:58 +11:00
sam 22aee72679 feat(judgement,approvals,governance,ui): a class earns a raise after N approvals and drops on denial or failure; a freeze holds changing actions (#2004) 2026-11-16 05:03:24 +11:00
sam a91df7ec1f feat(judgement,approvals,ui): action classes are records with a level, history and reset; the precedent guard reads the class level (#2003) 2026-11-16 04:35:12 +11:00
sam 0238f247e5 feat(approvals): a held action is asked of a person, and the run waits for the answer (archie #1985, #1999)
An action judgement holds at propose and wait is no longer a dead end. The
new approvals module owns the request: it binds to the decision and the
action's digest, so an answer covers the action the person was shown and
nothing else, and it settles once. The answer wakes the run, which judges the
action again and runs it only as the approval allows: a refusal still refuses,
an altered input runs nothing, and a judgement that came back worse runs
nothing on that approval.

The composer carries the request as a message with the request id and its
deadline, and its answer is a message too, so the trail shows what was asked
and what was answered. A request nobody answers inside its deadline is settled
as expired and nothing runs.

How long a request waits is an organisation setting, edited in the dashboard
and stored in milliseconds, because a Postgres interval scans as text. A run
pins the deadline it started with, like its model.

An agent that proposes the same action twice is asked about it once, and the
notification says when the investigation proposed nothing at all.
2026-11-16 02:21:06 +11:00
sam 0dbf241106 feat(judgement): the weakest-link decision, and executing only what was judged (#1986, #1989, #1990, #1988)
A new judgement module assesses one concrete action: hard rules refuse before
any guard runs; every guard sets a ceiling and the action runs at the lowest;
a guard that fails, panics or gives no ceiling holds it at propose and wait;
an advisory verdict is recorded and sets no ceiling. Reversibility and
Precedent (the fresh-install class default) are the first guards. Every
decision is recorded with its verdicts and class, and the run's trail lists
them. Agent tools now offer connector actions too: each call is assessed and
executed only as decided and only with the judged input, and a held action
comes back to the agent as a failed call naming its level and findings.
Tools gain Resolve, so an MCP server's declared effect reaches judgement.
2026-11-16 01:05:29 +11:00