Commit Graph
9 Commits
Author SHA1 Message Date
sam e127b2b05d feat(approvals): gate answers by role and link channel accounts through single-use codes, replacing the people model (#2095) 2026-11-22 02:13:08 +11:00
sam 4b494bf75b feat(approvals): add a People page, leave channel approvals unchanged, make a binding's target optional (#2072) 2026-11-21 21:37:42 +11:00
sam b9a60e03a3 feat(chat): bind held-action approvals to the whole action and require fresh passkey confirmation for irreversible ones (#2072) 2026-11-21 21:35:27 +11:00
sam bc72655c8e feat(judgement): refuse privileged and undeclared actions in chat (#2071)
Connector actions and lookups now declare privilege change and secret access,
and an MCP server declares both per installation. Chat withholds and refuses
any tool that has either effect or declared nothing, in judgement and at
execution, so an approval cannot convert the refusal. Workflows and the
dashboard are unchanged.
2026-11-21 21:28:06 +11:00
sam 073959b4f9 feat(memory): scope chat notes and preserve deletions on restore (#2069)
Keep personal notes private by default and require review before publishing shared notes. Restore backups and replay forgotten notes in one transaction, and retain the installed release when an unattended rollback could bring back forgotten notes.
2026-11-21 13:20:31 +11:00
sam 98c0c11cbb feat(chat): track session taint and hold mutations in tainted conversations (#2068)
A conversation gets a persisted tainted flag (migration 00040), set the
first time output from a tool whose declaration trust level is not
"trusted" enters the turn, before that output reaches the model. The
flag is only ever set, never cleared, so later turns and resumed
processes inherit it. In a tainted conversation every non-read-only
action is raised to a floor of propose_and_wait, so the existing hold
path renders the approval card with the exact judged action; read-only
lookups still run, and a hard-rule refusal stays a refusal. An
unreadable taint state or a failed taint write fails closed. Trust
levels live on connectors.Declaration; unknown levels are untrusted.
2026-11-21 03:15:04 +11:00
sam 96fb0afcd9 feat(chat): offer the model only the tools its person may use (#2067)
Discovery now applies the same per-tool policy decision as invocation: a connector lookup needs read, an action needs run, and each builtin carries the action and resource kind it checks. The model's tool schemas and the system prompt's summaries come from that one filtered set. A name that two declarations, or a declaration and a builtin, both claim is withheld from all of them.

The stream client fixes the wire tools for the whole turn, so a guarded provider re-reads the owner's access and the running declarations before every model request. If an offered tool is no longer permitted or declared, the turn stops before the request instead of describing a stale set. A call to a declaration the registry no longer holds is refused before anything runs, naming nothing.

Memory notes are left out of the prompt when the person may not read settings, so the tool filter is not bypassed by prompt context.
2026-11-21 01:45:03 +11:00
samandCopilot cf6108ad14 feat(chat): conversations belong to their person and act with their access (#2065)
Sessions carry an owner; every chat route serves only the caller's own
conversations (others are 404). Each turn and each approved held action
acts as its owner's principal, checked against the access rules before
judgement, so an operator's chat can't do what the operator can't. Chat is
open to any signed-in person, and the stream sends chat updates only to
the owner.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
2026-11-20 19:05:20 +11:00
samandCopilot cbfcd43018 feat(chat): add the chat module to the archie binary (#2018)
Conversations are the chat module's own tables. A turn is stored before it
is acknowledged, runs after the turns before it in its session, and records
everything it streams, so a reader replays it from a cursor. The agent uses
the configured agent model, the same connector tools as a run and the
dashboard's own reads; every action is judged, and one judgement holds is
put to people through approvals and run once when approved. A restart fails
the turns it interrupted visibly and never restarts a started action.

Conversation memory is a set of notes in the memory module. The chat
prompts are copied from internal/prompts (chat-system.tmpl, soul.md,
session-title.md) into internal/platform/prompts while the gateway still
reads them; the tool parameter redaction is copied from
internal/domain/messaging/tool_parameters.go.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
2026-11-19 19:40:44 +11:00