Commit Graph
442 Commits
Author SHA1 Message Date
sam bb0d7fa8d0 feat(archied): raise an alert when container max uptime is under the task limit 2026-10-26 19:48:09 +11:00
sam 8bcd89e050 fix(gateway): name a plan wait's workflow from its run, not the next one
workflow.approve pre-writes the task's workflow column with the workflow
that runs once approved, so task_list named a workflow that had not
started while the task's own log named the one that produced the plan.
Chat now names the compiled workflow through the pinned definition for a
plan wait, matching the log.
2026-10-26 18:32:25 +11:00
sam aadc86a2ca fix(workflow): report a parked run's outcome as the task's status
A run that parked left tc.Outcome empty, so the taskrun response carried an
empty status and the daemon logged 'taskrun run recorded no outcome' for a run
that had recorded one, then retried the park it could not apply.

Record the outcome once the parked transition lands, so a budget stop and every
other park surface with the reason the run parked.
2026-10-26 18:28:32 +11:00
sam 660e97f87e fix(taskactions): let the operator chat act on every task, like the dashboard
A chat profile bound to the operator identity passed a strict scope to the
action gate, so it refused root-owned tasks whose identity is the daemon's
empty sentinel (task 22: belongs to "", not "d898c8c8-..."). The dashboard
acts with a nil scope and the chat read paths already treat the operator as
acting across identities, so the write gate was the only surface that disagreed.

Exempt the operator identity from the scope comparison and share that rule with
the chat adapters that had their own copy.
2026-10-26 18:24:47 +11:00
sam 06d08f422b fix(workflow): record the own-PR flag after the review stages that allocate it
StagePRReviewAndOpenPR set tc.prReview.ownPR before the decision stages ran,
but intake allocates tc.prReview, so every implement run dereferenced a nil
state at repo.open-pr and the agent container panicked with exit code 2
before opening the pull request.
2026-10-26 18:07:01 +11:00
sam 8026c07319 feat(workflow): message.send step delivers text through archied's messaging client 2026-10-26 00:56:07 +11:00
sam cc5789eb5f feat(scheduling): schedules and runs in the State Store with grid timing and run history
Schedules leave the schedules control-plane resource for State Store tables
behind a ScheduleService contract. A run is claimed per schedule slot, missed
slots fire once, the next run stays on the schedule's grid, cron is read in the
schedule-settings time zone, and failed runs are recorded and advance. Three
failures in a row raise a schedule.failing alert. A fired run records a
schedule-fired capture. Run history is pruned by age and count.

Removes the chat and spawn-task runners, the dangerous-payload guard and the
old schedules settings page.
2026-10-26 00:49:24 +11:00
sam d1d6301fc2 feat(eda): capture schedule firings on a reserved source for bindings 2026-10-26 00:38:55 +11:00
sam 220302b41e fix(logging): route ui, messaging and playbook logs through the service log 2026-10-26 00:18:39 +11:00
sam 97655323b9 refactor(logging): log with the caller's context so records carry trace ids 2026-10-25 23:30:01 +11:00
sam 74b11b331d feat(natsrpc): propagate trace context across NATS and into task runs 2026-10-25 23:03:55 +11:00
sam 3912d7e3e0 fix(audit): harden export, scope history and derive feed targets 2026-10-25 22:39:22 +11:00
sam c95ff24f23 feat(webui): audit feed API, export and read_audit grant 2026-10-25 22:01:02 +11:00
sam 691002eb9b fix(state): harden the audit feed against audit findings 2026-10-25 21:23:52 +11:00
sam dd9b378431 feat(state): org-scoped audit feed contract and read model 2026-10-25 21:23:52 +11:00
sam f8c610ea5f fix(postgres): correct package and secret audit coverage 2026-10-25 21:03:00 +11:00
sam b397329de0 fix(state): correct audit backfill scoping and policy write races 2026-10-25 20:59:04 +11:00
sam 9bedaaefb4 feat(state): org-scoped audit trail and complete policy write audit 2026-10-25 20:49:20 +11:00
sam 2965f8354c fix(messaging): read a voice clip's container instead of believing its type
The dashboard upload names its own attachment type, so the check that ran on
it only compared two declarations the sender wrote. A browser can label any
bytes audio/ogg, and the transcriber keys on the attachment type alone, so the
bytes went to the operator's provider regardless.

When the bytes are present the container decides: Ogg, WebM/Matroska, mp4/m4a,
wav, flac and mp3 are the containers a clip is recorded in, and anything else
claimed to be speech is refused before it reaches a transcriber. Magic numbers
rather than http.DetectContentType, which reports application/octet-stream for
the WebM every browser but Safari records.
2026-10-25 20:17:30 +11:00
sam dda9aecde6 fix(messaging): say what the inbound media check actually checks
The check compares two declarations; it never reads the bytes, so it cannot
tell whether a file is what its content type says. The comment and the test
row claiming otherwise have gone, and so has an allowance letting a video
carry an image type, which no producer emits.

A caller that named the attachment type itself -- a browser upload -- must now
also state a content type, since the previous empty-type escape made the check
optional there. The Telegram declared-size rule is deleted: the shared check
runs first with the same ceiling, so only the rule on the bytes actually
received can still fire.
2026-10-25 18:13:30 +11:00
sam 147792451d fix(messaging): refuse an inbound attachment whose type contradicts its content
A declared MIME type decides what the Gateway does with a file: a voice note is
transcribed, an image goes to a model. Both acceptance points took that claim
from the caller unchecked -- the dashboard's voice upload and every Telegram
attachment -- so a file could be used as something it is not.

ValidateInbound refuses a type outside the shared vocabulary, content over the
size ceiling, and a MIME type that contradicts the attachment type, with an
error the caller sees: a dashboard upload is a 400, a Telegram file gets a
chat reply instead of a silent drop.
2026-10-25 18:05:34 +11:00
sam eabb7886f8 fix(workflow): review a task's own pull request as a comment
GitHub refuses REQUEST_CHANGES from a pull request's author, so a blocking
finding on the PR a task opened lost every finding. The retry's identical
refusal is now reported once, and forge errors drop the gRPC framing.
2026-10-25 17:57:20 +11:00
sam c265b07ec7 refactor(workflow): stop assuming a human decides
An approval, ask or park can be resolved by an operator, a rule, a
classifier or automatically, so the vocabulary stops naming a person:

- step type human.approve -> workflow.approve
- park class needs_human -> needs_decision (ParkNeedsDecision)
- event kinds human_* -> operator_*
- alert kind task.needs_human -> task.needs_decision
- status grouping "needs you" -> "needs attention"

0061 rewrites the stored values, including the approval step type in the
workflow-definitions resource and each task's pinned definition, with the
digest recomputed so resumed tasks still match.
2026-10-25 17:31:16 +11:00
sam 986fdabeaa docs: clarify reviewer in implement finish summary 2026-10-25 17:20:13 +11:00
sam 55ea1f68e5 docs(workflow): name no human approver in the plan document 2026-10-25 17:19:06 +11:00
sam ccef626118 feat(workflow): approve a PRD as one structured document
The prd step returns the PRD as structured fields and human.approve takes
them as its document. The plan text every surface reads and the approval
card's layout are both rendered from that one document, so what the human
approves is what the implement workflow is handed. Replaces the separately
generated OpenUI view, which could drift from the plan.
2026-10-25 16:53:17 +11:00
sam 056f767ff0 feat(curator): add and remove curators live; definition updates touch only changed curators 2026-10-25 16:45:34 +11:00
sam 181915313e feat(workflow): render a PRD's generated view on the plan approval card
agent.run gains views: the agent may return an OpenUI Lang view of its
answer through an optional view tool, read as steps.<id>.view. human.approve
stores it as the task's plan_view, and the approval card renders it with
the dashboard's own components, falling back to the plan text when there
is no view or it does not parse. The feasibility prd step uses it.
2026-10-25 16:33:04 +11:00
sam 5c922038f9 fix(access): check the policy version before each decision
archied, archie-ui and the tool gateway re-read the stored policies on a 30s
ticker, so a forbid written seconds earlier did not stop a run. One
instance-wide revision now moves inside the transaction of every policy write,
and a decision reads it first and rebuilds the engine only when it moved.

A version read that fails, or a policy list that fails after a bump, refuses
with ErrChainUnavailable instead of standing on the last chain, so three call
sites now treat unavailable as availability rather than denial: dispatch leaves
the capture for the next cycle, a task start parks transient, and the tool
gateway reports the tool as unavailable.
2026-10-25 16:21:01 +11:00
sam 64ab228658 feat(egress): record a test run's deliveries instead of sending them
A test run used to suppress only forge and push writes, which named one
destination among many. Deliveries now have one seam: TaskContext.Deliveries
records each as "would ..." and performs none, and a recording egress
session does the same for a Kit run's HTTP writes and central tool calls,
answering them synthetically. Provider hosts stay live, since model traffic
is how a test run reads. Native containers have no proxy (archie-core-n8d3).
2026-10-25 16:01:06 +11:00
sam 6f0ea530aa fix(archied): post each reportable status a chat task reaches, recorded by status not by flag 2026-10-25 02:42:31 +11:00
sam c7c1861ee7 refactor!: waiting_human becomes waiting_approval, with the approval kind on each task 2026-10-24 23:45:20 +11:00
sam d10e37cf66 fix(workflow): keep the failing lines when clipping gate output 2026-10-24 23:18:23 +11:00
sam 7abc816735 refactor!: remove the diff-size cap and its gate.diff-size step 2026-10-24 22:19:23 +11:00
sam 15a9bcebdf Revert "fix(messaging): mask secrets embedded in chat tool arguments"
This reverts commit 89221085d9.
2026-10-24 20:56:31 +11:00
sam 89221085d9 fix(messaging): mask secrets embedded in chat tool arguments 2026-10-24 20:53:27 +11:00
sam 4719b9cf00 feat(chat): show each tool call where it happened in the reply 2026-10-24 20:48:43 +11:00
sam 5e348b6ada fix(messaging): show tool argument values in chat, redacting only sensitive keys 2026-10-24 20:46:29 +11:00
sam c6c4f36a09 feat(gateway): record how long each chat tool call took and show it per row 2026-10-24 20:44:40 +11:00
sam 092fae3aa4 chore(workflow): say in a review post failure whether it carried inline comments 2026-10-24 18:27:17 +11:00
sam 24f1174d9f chore(workflow): log where a review's comments are placed before posting 2026-10-24 18:18:54 +11:00
sam 7338ad0e14 fix(workflow): repost a review without inline comments when the forge refuses their placement 2026-10-24 18:07:14 +11:00
sam 57c025688a fix(workflow): post review findings outside the diff in the review body instead of inline 2026-10-24 17:27:35 +11:00
sam 7a8414765c fix(daemon): park a pushing workflow before it runs when its identity may not push, with a link to the connection 2026-10-24 17:06:25 +11:00
sam 9b3ade37d3 fix(workflow): require triage's workflow when a code change is needed 2026-10-22 17:37:33 +11:00
sam 439fbeabc5 fix(triage): let a task that needs no code change be classified
The classify result required a workflow, and the capture tool rejects an empty
required string on the agent's single allowed call. A task needing no code
change has no workflow to name, so the classification failed instead of
closing the task -- the non-code path could never complete.

Require only what every classification has: whether a change is needed and the
reasons. Name the workflow required-when-true in its description, guard the
handoff on needs_code_change rather than relying on the finish step having
ended the run first, and comment the reasons before closing so the reporter
sees why nothing was built.
2026-10-22 13:50:47 +11:00
sam 2b431c43de feat(workflow): install a repository's declared tooling before its gate 2026-10-21 23:11:16 +11:00
sam 166969cdfb feat(chat): ask several questions in one interaction with ask_questions
A turn can pose a questionnaire (single and multiple choice, free text,
skippable) and read the answers back, rendered in the chat with the
shadcn-vue questionnaire. The gateway contract gains a questionnaire ask and a
can_question capability; answers travel as a JSON object keyed by question name
and are checked against the questions that were asked.
2026-10-19 03:30:47 +11:00
sam 65f2586121 feat(instructions): layer instance, organisation, workflow and repository instructions into every agent stage 2026-10-17 23:42:39 +11:00
sam 3f553bb6f2 feat(kit): central tools for harnesses through the egress proxy and gateway task-tool RPCs 2026-10-17 23:42:37 +11:00