mirror of
https://github.com/andrewyng/openworker.git
synced 2026-09-11 06:30:25 +00:00
Step 2: Auto-Approve mode - the reviewer, the hook, and the renames
The mode from ocw-context/docs/reviewed-auto-mode.md (rev. 4), v1 scope.
coworker/reviewer.py (new)
- The 8.3 prompt verbatim, cache-shaped: instructions + known world (folders
and remotes only) + user-message history in the stable prefix; this turn's
request and ONE action in the suffix.
- parse_verdict: any defect (empty, non-JSON, unknown verdict) -> unsure.
There is no parse path that results in execution (8.5).
- Reviewer.review never raises: provider errors and timeouts -> unsure.
Metering counters (checks / verdicts / tokens) for 1.7.
- AGENT_DENY_MESSAGE: the terse, non-diagnostic refusal the agent gets on a
deny; the full reason goes to the user only (8.4 asymmetry).
coworker/engine.py
- Reviewer consulted ONLY when: attached, mode is AUTO_APPROVE, session
explicitly attended (unset is_attended counts as NOT attended, so
automations can never be reviewed), fewer than two denials this turn.
- Consulted ONLY on decisions the gate marked needs_user - hard denies
never reach it, so it can only turn "ask" into "allow" (1.2).
- One action per request, fired concurrently for all of a turn's escalating
calls before the sequential authorize loop (8.6): a verdict cannot land
on the wrong action, and approval cards still reach the human one at a
time in call order.
- allow -> runs, audited with the reason. deny -> blocked; user event
carries the full reviewer reason + allow_anyway; agent message carries
only AGENT_DENY_MESSAGE. unsure -> today's card.
- Reviewer sees the user's words only, extracted mechanically from
role=user messages - never agent output, never tool results (4.4).
coworker/permissions.py
- Mode.AUTO renamed Mode.BYPASS_APPROVALS ("bypass-approvals"); legacy
"auto" still parses via _missing_ so configs, saved sessions, and the
golden decision table are untouched.
- Mode.AUTO_APPROVE ("auto-approve"): gate-identical to INTERACTIVE except
session grants ("always allow this ...") no longer auto-allow - they
route to the reviewer instead (1.5: out-of-band standing policy may skip
the judge; an in-flow click may not). Config allowlists still skip.
- _domain_allowed(include_session=False) checks the user-settings list only.
coworker/config.py: auto_approve flag, off by default, _GLOBAL_ONLY (a
cloned repo cannot hand itself a looser reviewer). agent.py attaches the
Reviewer only when the flag is on; without it AUTO_APPROVE behaves exactly
like INTERACTIVE.
server/manager.py: autonomy audit ranks auto-approve above interactive
(turning the reviewer on IS raising autonomy) and below bypass.
GUI: mode picker label "Full access" -> "Bypass approvals" (wire value
"auto" kept). Verified live against the real sidecar; e2e spec updated;
tsc and all 111 GUI unit tests pass.
Tests: tests/test_auto_approve.py (33) - gate behaviour per mode, fail-
closed parsing, prompt shape, deny asymmetry, retry guard, attended
gating, hard-deny isolation, per-action verdict landing, and that the
reviewer never sees agent prose. Permission suites + golden table: 146
passing unchanged.
This commit is contained in:
@@ -36,7 +36,7 @@ test("composer: send-gating, + attach menu, Mode menu", async ({ page }) => {
|
||||
await expect(menu.locator("button").filter({ hasText: "Ask for approval" })).toContainText("✓");
|
||||
await expect(menu.getByRole("switch", { name: "Send approvals to the Inbox" })).toBeVisible();
|
||||
// Picking an option closes the menu (and would flip the live engine's mode).
|
||||
await menu.getByText("Full access").click();
|
||||
await menu.getByText("Bypass approvals").click();
|
||||
await expect(page.getByTestId("mode-menu")).toHaveCount(0);
|
||||
});
|
||||
|
||||
|
||||
@@ -20,10 +20,14 @@ import {
|
||||
// polished enough to ship, and Custom (config.toml auto-allow rules) is a power-user mode
|
||||
// with no in-app explanation. The server still honors both — a session already in one of
|
||||
// those modes keeps working; the picker just doesn't offer them.
|
||||
// "auto" is the legacy wire value for Bypass approvals (server: Mode.BYPASS_APPROVALS) —
|
||||
// kept so saved sessions and configs keep working. Auto-Approve ("auto-approve") is the
|
||||
// reviewer mode (spec: reviewed-auto-mode.md); it appears only when the server says the
|
||||
// feature flag is on, wired in the settings pass — until then the picker omits it.
|
||||
const PERMISSION_OPTIONS: Option[] = [
|
||||
{ value: "discuss", label: "Discuss", description: "Chat and explore — no edits or commands" },
|
||||
{ value: "interactive", label: "Ask for approval", description: "Ask before edits and commands" },
|
||||
{ value: "auto", label: "Full access", description: "Run everything without asking" },
|
||||
{ value: "auto", label: "Bypass approvals", description: "Run everything without asking — approvals off" },
|
||||
];
|
||||
|
||||
// No hardcoded model fallback: until the server supplies the list (a few seconds after a
|
||||
|
||||
Reference in New Issue
Block a user