oh-my-codex 0.18.11 → 0.18.13
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/Cargo.lock +6 -6
- package/Cargo.toml +1 -1
- package/README.md +9 -1
- package/dist/autopilot/__tests__/ralplan-gate.test.js +668 -0
- package/dist/autopilot/__tests__/ralplan-gate.test.js.map +1 -1
- package/dist/autopilot/completion-gate.d.ts +10 -0
- package/dist/autopilot/completion-gate.d.ts.map +1 -0
- package/dist/autopilot/completion-gate.js +154 -0
- package/dist/autopilot/completion-gate.js.map +1 -0
- package/dist/autopilot/ralplan-gate.d.ts.map +1 -1
- package/dist/autopilot/ralplan-gate.js +42 -21
- package/dist/autopilot/ralplan-gate.js.map +1 -1
- package/dist/cli/__tests__/codex-plugin-layout.test.js +46 -3
- package/dist/cli/__tests__/codex-plugin-layout.test.js.map +1 -1
- package/dist/cli/__tests__/doctor-invalid-config.test.js +35 -0
- package/dist/cli/__tests__/doctor-invalid-config.test.js.map +1 -1
- package/dist/cli/__tests__/doctor-warning-copy.test.js +317 -0
- package/dist/cli/__tests__/doctor-warning-copy.test.js.map +1 -1
- package/dist/cli/__tests__/index.test.js +120 -2
- package/dist/cli/__tests__/index.test.js.map +1 -1
- package/dist/cli/__tests__/launch-fallback.test.js +1 -1
- package/dist/cli/__tests__/launch-fallback.test.js.map +1 -1
- package/dist/cli/__tests__/resume.test.js +217 -1
- package/dist/cli/__tests__/resume.test.js.map +1 -1
- package/dist/cli/__tests__/session-scoped-runtime.test.js +101 -0
- package/dist/cli/__tests__/session-scoped-runtime.test.js.map +1 -1
- package/dist/cli/__tests__/session-search-help.test.js +3 -2
- package/dist/cli/__tests__/session-search-help.test.js.map +1 -1
- package/dist/cli/__tests__/session-search.test.js +64 -2
- package/dist/cli/__tests__/session-search.test.js.map +1 -1
- package/dist/cli/__tests__/setup-agents-overwrite.test.js +289 -1
- package/dist/cli/__tests__/setup-agents-overwrite.test.js.map +1 -1
- package/dist/cli/__tests__/setup-install-mode.test.js +290 -17
- package/dist/cli/__tests__/setup-install-mode.test.js.map +1 -1
- package/dist/cli/__tests__/setup-prompts-overwrite.test.js +74 -0
- package/dist/cli/__tests__/setup-prompts-overwrite.test.js.map +1 -1
- package/dist/cli/__tests__/setup-scope.test.js +45 -0
- package/dist/cli/__tests__/setup-scope.test.js.map +1 -1
- package/dist/cli/__tests__/state.test.js +93 -0
- package/dist/cli/__tests__/state.test.js.map +1 -1
- package/dist/cli/__tests__/update.test.js +157 -3
- package/dist/cli/__tests__/update.test.js.map +1 -1
- package/dist/cli/__tests__/version-sync-contract.test.js +2 -0
- package/dist/cli/__tests__/version-sync-contract.test.js.map +1 -1
- package/dist/cli/doctor.d.ts.map +1 -1
- package/dist/cli/doctor.js +90 -12
- package/dist/cli/doctor.js.map +1 -1
- package/dist/cli/index.d.ts +13 -4
- package/dist/cli/index.d.ts.map +1 -1
- package/dist/cli/index.js +439 -46
- package/dist/cli/index.js.map +1 -1
- package/dist/cli/project-runtime-codex-homes.d.ts +6 -0
- package/dist/cli/project-runtime-codex-homes.d.ts.map +1 -0
- package/dist/cli/project-runtime-codex-homes.js +27 -0
- package/dist/cli/project-runtime-codex-homes.js.map +1 -0
- package/dist/cli/session-search.d.ts.map +1 -1
- package/dist/cli/session-search.js +8 -1
- package/dist/cli/session-search.js.map +1 -1
- package/dist/cli/setup.d.ts +2 -2
- package/dist/cli/setup.d.ts.map +1 -1
- package/dist/cli/setup.js +482 -126
- package/dist/cli/setup.js.map +1 -1
- package/dist/cli/state.d.ts.map +1 -1
- package/dist/cli/state.js +79 -8
- package/dist/cli/state.js.map +1 -1
- package/dist/cli/update.d.ts +1 -0
- package/dist/cli/update.d.ts.map +1 -1
- package/dist/cli/update.js +42 -10
- package/dist/cli/update.js.map +1 -1
- package/dist/config/__tests__/codex-hooks.test.js +73 -29
- package/dist/config/__tests__/codex-hooks.test.js.map +1 -1
- package/dist/config/codex-hooks.d.ts +14 -0
- package/dist/config/codex-hooks.d.ts.map +1 -1
- package/dist/config/codex-hooks.js +54 -51
- package/dist/config/codex-hooks.js.map +1 -1
- package/dist/config/generator.d.ts +1 -1
- package/dist/config/generator.d.ts.map +1 -1
- package/dist/config/generator.js +1 -1
- package/dist/config/generator.js.map +1 -1
- package/dist/hooks/__tests__/best-practice-research-skill.test.js +12 -0
- package/dist/hooks/__tests__/best-practice-research-skill.test.js.map +1 -1
- package/dist/hud/__tests__/authority.test.js +45 -12
- package/dist/hud/__tests__/authority.test.js.map +1 -1
- package/dist/hud/__tests__/reconcile.test.js +95 -0
- package/dist/hud/__tests__/reconcile.test.js.map +1 -1
- package/dist/hud/__tests__/render.test.js +6 -6
- package/dist/hud/__tests__/render.test.js.map +1 -1
- package/dist/hud/__tests__/tmux.test.js +2 -2
- package/dist/hud/__tests__/tmux.test.js.map +1 -1
- package/dist/hud/authority.d.ts.map +1 -1
- package/dist/hud/authority.js +17 -2
- package/dist/hud/authority.js.map +1 -1
- package/dist/hud/index.js +1 -4
- package/dist/hud/index.js.map +1 -1
- package/dist/hud/reconcile.d.ts.map +1 -1
- package/dist/hud/reconcile.js +42 -0
- package/dist/hud/reconcile.js.map +1 -1
- package/dist/hud/render.d.ts.map +1 -1
- package/dist/hud/render.js +6 -0
- package/dist/hud/render.js.map +1 -1
- package/dist/hud/tmux.d.ts.map +1 -1
- package/dist/hud/tmux.js +5 -4
- package/dist/hud/tmux.js.map +1 -1
- package/dist/mcp/__tests__/bootstrap.test.js +31 -1
- package/dist/mcp/__tests__/bootstrap.test.js.map +1 -1
- package/dist/mcp/bootstrap.d.ts +1 -0
- package/dist/mcp/bootstrap.d.ts.map +1 -1
- package/dist/mcp/bootstrap.js +32 -0
- package/dist/mcp/bootstrap.js.map +1 -1
- package/dist/modes/__tests__/base-autopilot-gates.test.d.ts +2 -0
- package/dist/modes/__tests__/base-autopilot-gates.test.d.ts.map +1 -0
- package/dist/modes/__tests__/base-autopilot-gates.test.js +154 -0
- package/dist/modes/__tests__/base-autopilot-gates.test.js.map +1 -0
- package/dist/modes/base.d.ts +4 -1
- package/dist/modes/base.d.ts.map +1 -1
- package/dist/modes/base.js +71 -1
- package/dist/modes/base.js.map +1 -1
- package/dist/pipeline/__tests__/orchestrator.test.js +144 -3
- package/dist/pipeline/__tests__/orchestrator.test.js.map +1 -1
- package/dist/pipeline/__tests__/stages.test.js +109 -0
- package/dist/pipeline/__tests__/stages.test.js.map +1 -1
- package/dist/pipeline/orchestrator.d.ts.map +1 -1
- package/dist/pipeline/orchestrator.js +11 -4
- package/dist/pipeline/orchestrator.js.map +1 -1
- package/dist/pipeline/stages/code-review.d.ts +2 -0
- package/dist/pipeline/stages/code-review.d.ts.map +1 -1
- package/dist/pipeline/stages/code-review.js +2 -0
- package/dist/pipeline/stages/code-review.js.map +1 -1
- package/dist/pipeline/stages/ultraqa.d.ts +3 -0
- package/dist/pipeline/stages/ultraqa.d.ts.map +1 -1
- package/dist/pipeline/stages/ultraqa.js +3 -0
- package/dist/pipeline/stages/ultraqa.js.map +1 -1
- package/dist/ralplan/__tests__/consensus-gate.test.d.ts +2 -0
- package/dist/ralplan/__tests__/consensus-gate.test.d.ts.map +1 -0
- package/dist/ralplan/__tests__/consensus-gate.test.js +631 -0
- package/dist/ralplan/__tests__/consensus-gate.test.js.map +1 -0
- package/dist/ralplan/consensus-gate.d.ts +9 -1
- package/dist/ralplan/consensus-gate.d.ts.map +1 -1
- package/dist/ralplan/consensus-gate.js +287 -65
- package/dist/ralplan/consensus-gate.js.map +1 -1
- package/dist/scripts/__tests__/codex-native-hook.test.js +481 -0
- package/dist/scripts/__tests__/codex-native-hook.test.js.map +1 -1
- package/dist/scripts/codex-native-hook.d.ts.map +1 -1
- package/dist/scripts/codex-native-hook.js +145 -25
- package/dist/scripts/codex-native-hook.js.map +1 -1
- package/dist/scripts/codex-native-pre-post.d.ts +1 -0
- package/dist/scripts/codex-native-pre-post.d.ts.map +1 -1
- package/dist/scripts/codex-native-pre-post.js +130 -0
- package/dist/scripts/codex-native-pre-post.js.map +1 -1
- package/dist/session-history/__tests__/search.test.js +166 -0
- package/dist/session-history/__tests__/search.test.js.map +1 -1
- package/dist/session-history/search.d.ts +7 -0
- package/dist/session-history/search.d.ts.map +1 -1
- package/dist/session-history/search.js +83 -24
- package/dist/session-history/search.js.map +1 -1
- package/dist/sidecar/__tests__/collector.test.js +60 -0
- package/dist/sidecar/__tests__/collector.test.js.map +1 -1
- package/dist/sidecar/collector.d.ts.map +1 -1
- package/dist/sidecar/collector.js +3 -6
- package/dist/sidecar/collector.js.map +1 -1
- package/dist/state/__tests__/operations.test.js +622 -0
- package/dist/state/__tests__/operations.test.js.map +1 -1
- package/dist/state/__tests__/skill-active.test.js +82 -0
- package/dist/state/__tests__/skill-active.test.js.map +1 -1
- package/dist/state/operations.d.ts.map +1 -1
- package/dist/state/operations.js +31 -9
- package/dist/state/operations.js.map +1 -1
- package/dist/state/skill-active.d.ts.map +1 -1
- package/dist/state/skill-active.js +41 -1
- package/dist/state/skill-active.js.map +1 -1
- package/dist/team/__tests__/runtime.test.js +81 -57
- package/dist/team/__tests__/runtime.test.js.map +1 -1
- package/dist/team/runtime.js +4 -4
- package/dist/team/runtime.js.map +1 -1
- package/dist/utils/__tests__/paths.test.js +23 -0
- package/dist/utils/__tests__/paths.test.js.map +1 -1
- package/dist/utils/__tests__/version.test.js +27 -0
- package/dist/utils/__tests__/version.test.js.map +1 -1
- package/dist/utils/paths.d.ts.map +1 -1
- package/dist/utils/paths.js +4 -2
- package/dist/utils/paths.js.map +1 -1
- package/dist/utils/version.d.ts.map +1 -1
- package/dist/utils/version.js +7 -2
- package/dist/utils/version.js.map +1 -1
- package/dist/verification/__tests__/ci-rust-gates.test.js +4 -2
- package/dist/verification/__tests__/ci-rust-gates.test.js.map +1 -1
- package/dist/verification/__tests__/dev-merge-issue-close-workflow.test.js +71 -3
- package/dist/verification/__tests__/dev-merge-issue-close-workflow.test.js.map +1 -1
- package/package.json +1 -1
- package/plugins/oh-my-codex/.codex-plugin/plugin.json +1 -1
- package/plugins/oh-my-codex/hooks/codex-native-hook.mjs +53 -2
- package/plugins/oh-my-codex/skills/best-practice-research/SKILL.md +6 -1
- package/skills/best-practice-research/SKILL.md +6 -1
- package/src/scripts/__tests__/codex-native-hook.test.ts +615 -0
- package/src/scripts/codex-native-hook.ts +162 -32
- package/src/scripts/codex-native-pre-post.ts +137 -0
|
@@ -6,6 +6,672 @@ import { tmpdir } from 'node:os';
|
|
|
6
6
|
import { subagentTrackingPath } from '../../subagents/tracker.js';
|
|
7
7
|
import { buildAutopilotRalplanUltragoalGateError, canAdvanceAutopilotRalplanToUltragoal, } from '../ralplan-gate.js';
|
|
8
8
|
describe('autopilot ralplan gate', () => {
|
|
9
|
+
it('rejects invalid next-state complete consensus before falling back to older valid current state', async () => {
|
|
10
|
+
const cwd = await mkdtemp(join(tmpdir(), 'omx-autopilot-ralplan-next-invalid-terminal-'));
|
|
11
|
+
const sessionId = 'sess-autopilot-next-invalid-terminal';
|
|
12
|
+
const trackingPath = subagentTrackingPath(cwd);
|
|
13
|
+
try {
|
|
14
|
+
await mkdir(join(trackingPath, '..'), { recursive: true });
|
|
15
|
+
await writeFile(trackingPath, JSON.stringify({
|
|
16
|
+
schemaVersion: 1,
|
|
17
|
+
sessions: {
|
|
18
|
+
[sessionId]: {
|
|
19
|
+
session_id: sessionId,
|
|
20
|
+
leader_thread_id: 'thread-leader',
|
|
21
|
+
updated_at: '2026-06-12T10:00:00.000Z',
|
|
22
|
+
threads: {
|
|
23
|
+
'thread-leader': { thread_id: 'thread-leader', kind: 'leader', first_seen_at: '2026-06-12T09:59:00.000Z', last_seen_at: '2026-06-12T09:59:00.000Z', turn_count: 1 },
|
|
24
|
+
'thread-architect-old': { thread_id: 'thread-architect-old', kind: 'subagent', first_seen_at: '2026-06-12T09:59:30.000Z', last_seen_at: '2026-06-12T09:59:30.000Z', completed_at: '2026-06-12T09:59:30.000Z', turn_count: 1 },
|
|
25
|
+
'thread-critic-old': { thread_id: 'thread-critic-old', kind: 'subagent', first_seen_at: '2026-06-12T10:00:00.000Z', last_seen_at: '2026-06-12T10:00:00.000Z', completed_at: '2026-06-12T10:00:00.000Z', turn_count: 1 },
|
|
26
|
+
},
|
|
27
|
+
},
|
|
28
|
+
},
|
|
29
|
+
}, null, 2));
|
|
30
|
+
const nextState = {
|
|
31
|
+
current_phase: 'ralplan',
|
|
32
|
+
return_to_ralplan_reason: 'Code review requested a plan update.',
|
|
33
|
+
handoff_artifacts: {
|
|
34
|
+
ralplan_consensus_gate: {
|
|
35
|
+
complete: true,
|
|
36
|
+
sequence: ['architect-review', 'critic-review'],
|
|
37
|
+
ralplan_architect_review: {
|
|
38
|
+
agent_role: 'architect',
|
|
39
|
+
provenance_kind: 'native_subagent',
|
|
40
|
+
verdict: 'iterate',
|
|
41
|
+
session_id: sessionId,
|
|
42
|
+
thread_id: 'thread-architect-new',
|
|
43
|
+
artifact_path: '.omx/artifacts/architect-new.md',
|
|
44
|
+
tracker_path: '.omx/state/subagent-tracking.json',
|
|
45
|
+
completed_at: '2026-06-12T10:01:00.000Z',
|
|
46
|
+
},
|
|
47
|
+
ralplan_critic_review: {
|
|
48
|
+
agent_role: 'critic',
|
|
49
|
+
provenance_kind: 'native_subagent',
|
|
50
|
+
verdict: 'approve',
|
|
51
|
+
session_id: sessionId,
|
|
52
|
+
thread_id: 'thread-critic-new',
|
|
53
|
+
artifact_path: '.omx/artifacts/critic-new.md',
|
|
54
|
+
tracker_path: '.omx/state/subagent-tracking.json',
|
|
55
|
+
completed_at: '2026-06-12T10:02:00.000Z',
|
|
56
|
+
},
|
|
57
|
+
},
|
|
58
|
+
},
|
|
59
|
+
};
|
|
60
|
+
const currentState = {
|
|
61
|
+
current_phase: 'ralplan',
|
|
62
|
+
handoff_artifacts: {
|
|
63
|
+
ralplan_consensus_gate: {
|
|
64
|
+
complete: true,
|
|
65
|
+
sequence: ['architect-review', 'critic-review'],
|
|
66
|
+
ralplan_architect_review: {
|
|
67
|
+
agent_role: 'architect',
|
|
68
|
+
provenance_kind: 'native_subagent',
|
|
69
|
+
verdict: 'approve',
|
|
70
|
+
session_id: sessionId,
|
|
71
|
+
thread_id: 'thread-architect-old',
|
|
72
|
+
artifact_path: '.omx/artifacts/architect-old.md',
|
|
73
|
+
tracker_path: '.omx/state/subagent-tracking.json',
|
|
74
|
+
completed_at: '2026-06-12T09:59:30.000Z',
|
|
75
|
+
},
|
|
76
|
+
ralplan_critic_review: {
|
|
77
|
+
agent_role: 'critic',
|
|
78
|
+
provenance_kind: 'native_subagent',
|
|
79
|
+
verdict: 'approve',
|
|
80
|
+
session_id: sessionId,
|
|
81
|
+
thread_id: 'thread-critic-old',
|
|
82
|
+
artifact_path: '.omx/artifacts/critic-old.md',
|
|
83
|
+
tracker_path: '.omx/state/subagent-tracking.json',
|
|
84
|
+
completed_at: '2026-06-12T10:00:00.000Z',
|
|
85
|
+
},
|
|
86
|
+
},
|
|
87
|
+
},
|
|
88
|
+
};
|
|
89
|
+
const decision = canAdvanceAutopilotRalplanToUltragoal({ cwd, sessionId, nextState, currentState });
|
|
90
|
+
assert.equal(decision.allowed, false);
|
|
91
|
+
assert.equal(decision.evidence?.source, 'next-autopilot-state:handoff_artifacts');
|
|
92
|
+
assert.equal(decision.evidence?.blockedReason, 'non_approving_ralplan_consensus_review');
|
|
93
|
+
assert.match(buildAutopilotRalplanUltragoalGateError(decision), /architect.*verdict=iterate/i);
|
|
94
|
+
}
|
|
95
|
+
finally {
|
|
96
|
+
await rm(cwd, { recursive: true, force: true });
|
|
97
|
+
}
|
|
98
|
+
});
|
|
99
|
+
it('rejects stale nested ralplan handoff consensus when parent state returned to ralplan', async () => {
|
|
100
|
+
const cwd = await mkdtemp(join(tmpdir(), 'omx-autopilot-ralplan-stale-nested-'));
|
|
101
|
+
const sessionId = 'sess-autopilot-stale-nested';
|
|
102
|
+
const trackingPath = subagentTrackingPath(cwd);
|
|
103
|
+
try {
|
|
104
|
+
await mkdir(join(trackingPath, '..'), { recursive: true });
|
|
105
|
+
await writeFile(trackingPath, JSON.stringify({
|
|
106
|
+
schemaVersion: 1,
|
|
107
|
+
sessions: {
|
|
108
|
+
[sessionId]: {
|
|
109
|
+
session_id: sessionId,
|
|
110
|
+
leader_thread_id: 'thread-leader',
|
|
111
|
+
updated_at: '2026-06-12T10:00:00.000Z',
|
|
112
|
+
threads: {
|
|
113
|
+
'thread-leader': { thread_id: 'thread-leader', kind: 'leader', first_seen_at: '2026-06-12T09:59:00.000Z', last_seen_at: '2026-06-12T09:59:00.000Z', turn_count: 1 },
|
|
114
|
+
'thread-architect-stale': { thread_id: 'thread-architect-stale', kind: 'subagent', first_seen_at: '2026-06-12T09:59:30.000Z', last_seen_at: '2026-06-12T09:59:30.000Z', completed_at: '2026-06-12T09:59:30.000Z', turn_count: 1 },
|
|
115
|
+
'thread-critic-stale': { thread_id: 'thread-critic-stale', kind: 'subagent', first_seen_at: '2026-06-12T10:00:00.000Z', last_seen_at: '2026-06-12T10:00:00.000Z', completed_at: '2026-06-12T10:00:00.000Z', turn_count: 1 },
|
|
116
|
+
},
|
|
117
|
+
},
|
|
118
|
+
},
|
|
119
|
+
}, null, 2));
|
|
120
|
+
const nextState = {
|
|
121
|
+
current_phase: 'ralplan',
|
|
122
|
+
return_to_ralplan_reason: 'Code review requested a plan update.',
|
|
123
|
+
review_cycle: 2,
|
|
124
|
+
handoff_artifacts: {
|
|
125
|
+
ralplan: {
|
|
126
|
+
review_cycle: 2,
|
|
127
|
+
ralplanConsensusGate: {
|
|
128
|
+
complete: true,
|
|
129
|
+
sequence: ['architect-review', 'critic-review'],
|
|
130
|
+
ralplan_architect_review: {
|
|
131
|
+
agent_role: 'architect',
|
|
132
|
+
provenance_kind: 'native_subagent',
|
|
133
|
+
verdict: 'approve',
|
|
134
|
+
session_id: sessionId,
|
|
135
|
+
thread_id: 'thread-architect-stale',
|
|
136
|
+
artifact_path: '.omx/artifacts/architect-stale.md',
|
|
137
|
+
tracker_path: '.omx/state/subagent-tracking.json',
|
|
138
|
+
completed_at: '2026-06-12T09:59:30.000Z',
|
|
139
|
+
},
|
|
140
|
+
ralplan_critic_review: {
|
|
141
|
+
agent_role: 'critic',
|
|
142
|
+
provenance_kind: 'native_subagent',
|
|
143
|
+
verdict: 'approve',
|
|
144
|
+
session_id: sessionId,
|
|
145
|
+
thread_id: 'thread-critic-stale',
|
|
146
|
+
artifact_path: '.omx/artifacts/critic-stale.md',
|
|
147
|
+
tracker_path: '.omx/state/subagent-tracking.json',
|
|
148
|
+
completed_at: '2026-06-12T10:00:00.000Z',
|
|
149
|
+
},
|
|
150
|
+
},
|
|
151
|
+
},
|
|
152
|
+
},
|
|
153
|
+
};
|
|
154
|
+
const decision = canAdvanceAutopilotRalplanToUltragoal({ cwd, sessionId, nextState });
|
|
155
|
+
assert.equal(decision.allowed, false);
|
|
156
|
+
assert.equal(decision.evidence?.blockedReason, 'missing_sequential_architect_then_critic_approval');
|
|
157
|
+
}
|
|
158
|
+
finally {
|
|
159
|
+
await rm(cwd, { recursive: true, force: true });
|
|
160
|
+
}
|
|
161
|
+
});
|
|
162
|
+
it('rejects stale raw handoff consensus when parent state returned to ralplan', async () => {
|
|
163
|
+
const cwd = await mkdtemp(join(tmpdir(), 'omx-autopilot-ralplan-stale-raw-handoff-'));
|
|
164
|
+
const sessionId = 'sess-autopilot-stale-raw-handoff';
|
|
165
|
+
const trackingPath = subagentTrackingPath(cwd);
|
|
166
|
+
try {
|
|
167
|
+
await mkdir(join(trackingPath, '..'), { recursive: true });
|
|
168
|
+
await writeFile(trackingPath, JSON.stringify({
|
|
169
|
+
schemaVersion: 1,
|
|
170
|
+
sessions: {
|
|
171
|
+
[sessionId]: {
|
|
172
|
+
session_id: sessionId,
|
|
173
|
+
leader_thread_id: 'thread-leader',
|
|
174
|
+
updated_at: '2026-06-12T10:00:00.000Z',
|
|
175
|
+
threads: {
|
|
176
|
+
'thread-leader': { thread_id: 'thread-leader', kind: 'leader', first_seen_at: '2026-06-12T09:59:00.000Z', last_seen_at: '2026-06-12T09:59:00.000Z', turn_count: 1 },
|
|
177
|
+
'thread-architect-stale': { thread_id: 'thread-architect-stale', kind: 'subagent', first_seen_at: '2026-06-12T09:59:30.000Z', last_seen_at: '2026-06-12T09:59:30.000Z', completed_at: '2026-06-12T09:59:30.000Z', turn_count: 1 },
|
|
178
|
+
'thread-critic-stale': { thread_id: 'thread-critic-stale', kind: 'subagent', first_seen_at: '2026-06-12T10:00:00.000Z', last_seen_at: '2026-06-12T10:00:00.000Z', completed_at: '2026-06-12T10:00:00.000Z', turn_count: 1 },
|
|
179
|
+
},
|
|
180
|
+
},
|
|
181
|
+
},
|
|
182
|
+
}, null, 2));
|
|
183
|
+
const currentState = {
|
|
184
|
+
current_phase: 'ralplan',
|
|
185
|
+
return_to_ralplan_reason: 'Code review requested a plan update.',
|
|
186
|
+
review_cycle: 1,
|
|
187
|
+
handoff_artifacts: {
|
|
188
|
+
ralplan_consensus_gate: {
|
|
189
|
+
complete: true,
|
|
190
|
+
sequence: ['architect-review', 'critic-review'],
|
|
191
|
+
ralplan_architect_review: {
|
|
192
|
+
agent_role: 'architect',
|
|
193
|
+
provenance_kind: 'native_subagent',
|
|
194
|
+
verdict: 'approve',
|
|
195
|
+
session_id: sessionId,
|
|
196
|
+
thread_id: 'thread-architect-stale',
|
|
197
|
+
artifact_path: '.omx/artifacts/architect-stale.md',
|
|
198
|
+
tracker_path: '.omx/state/subagent-tracking.json',
|
|
199
|
+
completed_at: '2026-06-12T09:59:30.000Z',
|
|
200
|
+
},
|
|
201
|
+
ralplan_critic_review: {
|
|
202
|
+
agent_role: 'critic',
|
|
203
|
+
provenance_kind: 'native_subagent',
|
|
204
|
+
verdict: 'approve',
|
|
205
|
+
session_id: sessionId,
|
|
206
|
+
thread_id: 'thread-critic-stale',
|
|
207
|
+
artifact_path: '.omx/artifacts/critic-stale.md',
|
|
208
|
+
tracker_path: '.omx/state/subagent-tracking.json',
|
|
209
|
+
completed_at: '2026-06-12T10:00:00.000Z',
|
|
210
|
+
},
|
|
211
|
+
},
|
|
212
|
+
},
|
|
213
|
+
};
|
|
214
|
+
const decision = canAdvanceAutopilotRalplanToUltragoal({ cwd, sessionId, currentState });
|
|
215
|
+
assert.equal(decision.allowed, false);
|
|
216
|
+
assert.equal(decision.evidence?.blockedReason, 'missing_sequential_architect_then_critic_approval');
|
|
217
|
+
}
|
|
218
|
+
finally {
|
|
219
|
+
await rm(cwd, { recursive: true, force: true });
|
|
220
|
+
}
|
|
221
|
+
});
|
|
222
|
+
for (const { lane, architectVerdict, criticVerdict } of [
|
|
223
|
+
{ lane: 'architect', architectVerdict: 'iterate', criticVerdict: 'approve' },
|
|
224
|
+
{ lane: 'critic', architectVerdict: 'approve', criticVerdict: 'iterate' },
|
|
225
|
+
]) {
|
|
226
|
+
it(`rejects complete ralplan consensus when ${lane} review verdict is iterate`, () => {
|
|
227
|
+
const state = {
|
|
228
|
+
current_phase: 'ralplan',
|
|
229
|
+
handoff_artifacts: {
|
|
230
|
+
ralplan_consensus_gate: {
|
|
231
|
+
complete: true,
|
|
232
|
+
sequence: ['architect-review', 'critic-review'],
|
|
233
|
+
ralplan_architect_review: {
|
|
234
|
+
agent_role: 'architect',
|
|
235
|
+
provenance_kind: 'native_subagent',
|
|
236
|
+
verdict: architectVerdict,
|
|
237
|
+
session_id: 'sess-autopilot-iterate',
|
|
238
|
+
thread_id: 'thread-architect',
|
|
239
|
+
artifact_path: '.omx/artifacts/architect.md',
|
|
240
|
+
tracker_path: '.omx/state/subagent-tracking.json',
|
|
241
|
+
completed_at: '2026-05-28T18:34:51.000Z',
|
|
242
|
+
},
|
|
243
|
+
ralplan_critic_review: {
|
|
244
|
+
agent_role: 'critic',
|
|
245
|
+
provenance_kind: 'native_subagent',
|
|
246
|
+
verdict: criticVerdict,
|
|
247
|
+
session_id: 'sess-autopilot-iterate',
|
|
248
|
+
thread_id: 'thread-critic',
|
|
249
|
+
artifact_path: '.omx/artifacts/critic.md',
|
|
250
|
+
tracker_path: '.omx/state/subagent-tracking.json',
|
|
251
|
+
completed_at: '2026-05-28T18:35:10.000Z',
|
|
252
|
+
},
|
|
253
|
+
},
|
|
254
|
+
},
|
|
255
|
+
};
|
|
256
|
+
const decision = canAdvanceAutopilotRalplanToUltragoal({
|
|
257
|
+
cwd: process.cwd(),
|
|
258
|
+
sessionId: 'sess-autopilot-iterate',
|
|
259
|
+
currentState: state,
|
|
260
|
+
});
|
|
261
|
+
assert.equal(decision.allowed, false);
|
|
262
|
+
assert.equal(decision.evidence?.blockedReason, 'non_approving_ralplan_consensus_review');
|
|
263
|
+
const error = buildAutopilotRalplanUltragoalGateError(decision);
|
|
264
|
+
assert.match(error, /non-approving architect or critic review evidence/i);
|
|
265
|
+
assert.doesNotMatch(error, /missing ralplan consensus gate/i);
|
|
266
|
+
assert.match(error, new RegExp(`${lane}.*verdict=iterate`, 'i'));
|
|
267
|
+
});
|
|
268
|
+
}
|
|
269
|
+
it('accepts fresh next-state consensus over stale invalid current-state consensus', async () => {
|
|
270
|
+
const cwd = await mkdtemp(join(tmpdir(), 'omx-autopilot-ralplan-fresh-next-valid-'));
|
|
271
|
+
const sessionId = 'sess-autopilot-fresh-next-valid';
|
|
272
|
+
const trackingPath = subagentTrackingPath(cwd);
|
|
273
|
+
try {
|
|
274
|
+
await mkdir(join(trackingPath, '..'), { recursive: true });
|
|
275
|
+
await writeFile(trackingPath, JSON.stringify({
|
|
276
|
+
schemaVersion: 1,
|
|
277
|
+
sessions: {
|
|
278
|
+
[sessionId]: {
|
|
279
|
+
session_id: sessionId,
|
|
280
|
+
leader_thread_id: 'thread-leader',
|
|
281
|
+
updated_at: '2026-06-12T10:03:00.000Z',
|
|
282
|
+
threads: {
|
|
283
|
+
'thread-leader': { thread_id: 'thread-leader', kind: 'leader', first_seen_at: '2026-06-12T09:59:00.000Z', last_seen_at: '2026-06-12T09:59:00.000Z', turn_count: 1 },
|
|
284
|
+
'thread-architect-fresh': { thread_id: 'thread-architect-fresh', kind: 'subagent', first_seen_at: '2026-06-12T10:02:00.000Z', last_seen_at: '2026-06-12T10:02:00.000Z', completed_at: '2026-06-12T10:02:00.000Z', turn_count: 1 },
|
|
285
|
+
'thread-critic-fresh': { thread_id: 'thread-critic-fresh', kind: 'subagent', first_seen_at: '2026-06-12T10:03:00.000Z', last_seen_at: '2026-06-12T10:03:00.000Z', completed_at: '2026-06-12T10:03:00.000Z', turn_count: 1 },
|
|
286
|
+
},
|
|
287
|
+
},
|
|
288
|
+
},
|
|
289
|
+
}, null, 2));
|
|
290
|
+
const nextState = {
|
|
291
|
+
current_phase: 'ralplan',
|
|
292
|
+
review_cycle: 2,
|
|
293
|
+
handoff_artifacts: {
|
|
294
|
+
ralplan_consensus_gate: {
|
|
295
|
+
complete: true,
|
|
296
|
+
sequence: ['architect-review', 'critic-review'],
|
|
297
|
+
ralplan_architect_review: {
|
|
298
|
+
agent_role: 'architect',
|
|
299
|
+
provenance_kind: 'native_subagent',
|
|
300
|
+
verdict: 'approve',
|
|
301
|
+
review_cycle: 2,
|
|
302
|
+
session_id: sessionId,
|
|
303
|
+
thread_id: 'thread-architect-fresh',
|
|
304
|
+
artifact_path: '.omx/artifacts/architect-fresh.md',
|
|
305
|
+
tracker_path: '.omx/state/subagent-tracking.json',
|
|
306
|
+
completed_at: '2026-06-12T10:02:00.000Z',
|
|
307
|
+
},
|
|
308
|
+
ralplan_critic_review: {
|
|
309
|
+
agent_role: 'critic',
|
|
310
|
+
provenance_kind: 'native_subagent',
|
|
311
|
+
verdict: 'approve',
|
|
312
|
+
review_cycle: 2,
|
|
313
|
+
session_id: sessionId,
|
|
314
|
+
thread_id: 'thread-critic-fresh',
|
|
315
|
+
artifact_path: '.omx/artifacts/critic-fresh.md',
|
|
316
|
+
tracker_path: '.omx/state/subagent-tracking.json',
|
|
317
|
+
completed_at: '2026-06-12T10:03:00.000Z',
|
|
318
|
+
},
|
|
319
|
+
},
|
|
320
|
+
},
|
|
321
|
+
};
|
|
322
|
+
const currentState = {
|
|
323
|
+
current_phase: 'ralplan',
|
|
324
|
+
return_to_ralplan_reason: 'Code review requested a plan update.',
|
|
325
|
+
review_cycle: 1,
|
|
326
|
+
handoff_artifacts: {
|
|
327
|
+
ralplan_consensus_gate: {
|
|
328
|
+
complete: true,
|
|
329
|
+
sequence: ['architect-review', 'critic-review'],
|
|
330
|
+
ralplan_architect_review: {
|
|
331
|
+
agent_role: 'architect',
|
|
332
|
+
provenance_kind: 'native_subagent',
|
|
333
|
+
verdict: 'block',
|
|
334
|
+
review_cycle: 1,
|
|
335
|
+
session_id: sessionId,
|
|
336
|
+
thread_id: 'thread-architect-stale',
|
|
337
|
+
artifact_path: '.omx/artifacts/architect-stale.md',
|
|
338
|
+
tracker_path: '.omx/state/subagent-tracking.json',
|
|
339
|
+
completed_at: '2026-06-12T10:00:00.000Z',
|
|
340
|
+
},
|
|
341
|
+
ralplan_critic_review: {
|
|
342
|
+
agent_role: 'critic',
|
|
343
|
+
provenance_kind: 'native_subagent',
|
|
344
|
+
verdict: 'approve',
|
|
345
|
+
review_cycle: 1,
|
|
346
|
+
session_id: sessionId,
|
|
347
|
+
thread_id: 'thread-critic-stale',
|
|
348
|
+
artifact_path: '.omx/artifacts/critic-stale.md',
|
|
349
|
+
tracker_path: '.omx/state/subagent-tracking.json',
|
|
350
|
+
completed_at: '2026-06-12T10:01:00.000Z',
|
|
351
|
+
},
|
|
352
|
+
},
|
|
353
|
+
},
|
|
354
|
+
};
|
|
355
|
+
const decision = canAdvanceAutopilotRalplanToUltragoal({ cwd, sessionId, nextState, currentState });
|
|
356
|
+
assert.equal(decision.allowed, true);
|
|
357
|
+
assert.equal(decision.evidence?.source, 'next-autopilot-state');
|
|
358
|
+
assert.equal(decision.evidence?.blockedReason, null);
|
|
359
|
+
}
|
|
360
|
+
finally {
|
|
361
|
+
await rm(cwd, { recursive: true, force: true });
|
|
362
|
+
}
|
|
363
|
+
});
|
|
364
|
+
it('accepts tracker-backed native reviews without duplicated session, tracker, or artifact fields', async () => {
|
|
365
|
+
const cwd = await mkdtemp(join(tmpdir(), 'omx-autopilot-ralplan-tracker-resolved-'));
|
|
366
|
+
const sessionId = 'sess-autopilot-tracker-resolved';
|
|
367
|
+
const trackingPath = subagentTrackingPath(cwd);
|
|
368
|
+
try {
|
|
369
|
+
await mkdir(join(trackingPath, '..'), { recursive: true });
|
|
370
|
+
await writeFile(trackingPath, JSON.stringify({
|
|
371
|
+
schemaVersion: 1,
|
|
372
|
+
sessions: {
|
|
373
|
+
[sessionId]: {
|
|
374
|
+
session_id: sessionId,
|
|
375
|
+
leader_thread_id: 'thread-leader',
|
|
376
|
+
updated_at: '2026-06-12T10:03:00.000Z',
|
|
377
|
+
threads: {
|
|
378
|
+
'thread-leader': { thread_id: 'thread-leader', kind: 'leader', first_seen_at: '2026-06-12T09:59:00.000Z', last_seen_at: '2026-06-12T09:59:00.000Z', turn_count: 1 },
|
|
379
|
+
'thread-architect': { thread_id: 'thread-architect', kind: 'subagent', first_seen_at: '2026-06-12T10:02:00.000Z', last_seen_at: '2026-06-12T10:02:00.000Z', completed_at: '2026-06-12T10:02:00.000Z', turn_count: 1 },
|
|
380
|
+
'thread-critic': { thread_id: 'thread-critic', kind: 'subagent', first_seen_at: '2026-06-12T10:03:00.000Z', last_seen_at: '2026-06-12T10:03:00.000Z', completed_at: '2026-06-12T10:03:00.000Z', turn_count: 1 },
|
|
381
|
+
},
|
|
382
|
+
},
|
|
383
|
+
},
|
|
384
|
+
}, null, 2));
|
|
385
|
+
const state = {
|
|
386
|
+
current_phase: 'ralplan',
|
|
387
|
+
handoff_artifacts: {
|
|
388
|
+
ralplan_consensus_gate: {
|
|
389
|
+
complete: true,
|
|
390
|
+
sequence: ['architect-review', 'critic-review'],
|
|
391
|
+
ralplan_architect_review: {
|
|
392
|
+
agent_role: 'architect',
|
|
393
|
+
provenance_kind: 'native_subagent',
|
|
394
|
+
verdict: 'approve',
|
|
395
|
+
thread_id: 'thread-architect',
|
|
396
|
+
completed_at: '2026-06-12T10:02:00.000Z',
|
|
397
|
+
},
|
|
398
|
+
ralplan_critic_review: {
|
|
399
|
+
agent_role: 'critic',
|
|
400
|
+
provenance_kind: 'native_subagent',
|
|
401
|
+
verdict: 'approve',
|
|
402
|
+
thread_id: 'thread-critic',
|
|
403
|
+
completed_at: '2026-06-12T10:03:00.000Z',
|
|
404
|
+
},
|
|
405
|
+
},
|
|
406
|
+
},
|
|
407
|
+
};
|
|
408
|
+
const decision = canAdvanceAutopilotRalplanToUltragoal({ cwd, sessionId, currentState: state });
|
|
409
|
+
assert.equal(decision.allowed, true);
|
|
410
|
+
assert.equal(decision.evidence?.blockedReason, null);
|
|
411
|
+
}
|
|
412
|
+
finally {
|
|
413
|
+
await rm(cwd, { recursive: true, force: true });
|
|
414
|
+
}
|
|
415
|
+
});
|
|
416
|
+
it('rejects omitted review session ids when no transition session context exists', async () => {
|
|
417
|
+
const cwd = await mkdtemp(join(tmpdir(), 'omx-autopilot-ralplan-no-session-context-'));
|
|
418
|
+
const trackingPath = subagentTrackingPath(cwd);
|
|
419
|
+
try {
|
|
420
|
+
await mkdir(join(trackingPath, '..'), { recursive: true });
|
|
421
|
+
await writeFile(trackingPath, JSON.stringify({
|
|
422
|
+
schemaVersion: 1,
|
|
423
|
+
sessions: {},
|
|
424
|
+
}, null, 2));
|
|
425
|
+
const state = {
|
|
426
|
+
current_phase: 'ralplan',
|
|
427
|
+
handoff_artifacts: {
|
|
428
|
+
ralplan_consensus_gate: {
|
|
429
|
+
complete: true,
|
|
430
|
+
sequence: ['architect-review', 'critic-review'],
|
|
431
|
+
ralplan_architect_review: {
|
|
432
|
+
agent_role: 'architect',
|
|
433
|
+
provenance_kind: 'native_subagent',
|
|
434
|
+
verdict: 'approve',
|
|
435
|
+
thread_id: 'thread-architect',
|
|
436
|
+
},
|
|
437
|
+
ralplan_critic_review: {
|
|
438
|
+
agent_role: 'critic',
|
|
439
|
+
provenance_kind: 'native_subagent',
|
|
440
|
+
verdict: 'approve',
|
|
441
|
+
thread_id: 'thread-critic',
|
|
442
|
+
},
|
|
443
|
+
},
|
|
444
|
+
},
|
|
445
|
+
};
|
|
446
|
+
const decision = canAdvanceAutopilotRalplanToUltragoal({ cwd, currentState: state });
|
|
447
|
+
assert.equal(decision.allowed, false);
|
|
448
|
+
assert.match(buildAutopilotRalplanUltragoalGateError(decision), /cannot resolve session_id/);
|
|
449
|
+
}
|
|
450
|
+
finally {
|
|
451
|
+
await rm(cwd, { recursive: true, force: true });
|
|
452
|
+
}
|
|
453
|
+
});
|
|
454
|
+
it('rejects architect and critic reviews that reuse the same native tracker thread', async () => {
|
|
455
|
+
const cwd = await mkdtemp(join(tmpdir(), 'omx-autopilot-ralplan-duplicate-thread-'));
|
|
456
|
+
const sessionId = 'sess-autopilot-duplicate-thread';
|
|
457
|
+
const trackingPath = subagentTrackingPath(cwd);
|
|
458
|
+
try {
|
|
459
|
+
await mkdir(join(trackingPath, '..'), { recursive: true });
|
|
460
|
+
await writeFile(trackingPath, JSON.stringify({
|
|
461
|
+
schemaVersion: 1,
|
|
462
|
+
sessions: {
|
|
463
|
+
[sessionId]: {
|
|
464
|
+
session_id: sessionId,
|
|
465
|
+
leader_thread_id: 'thread-leader',
|
|
466
|
+
updated_at: '2026-06-12T10:03:00.000Z',
|
|
467
|
+
threads: {
|
|
468
|
+
'thread-leader': { thread_id: 'thread-leader', kind: 'leader', first_seen_at: '2026-06-12T09:59:00.000Z', last_seen_at: '2026-06-12T09:59:00.000Z', turn_count: 1 },
|
|
469
|
+
'thread-reviewer': { thread_id: 'thread-reviewer', kind: 'subagent', first_seen_at: '2026-06-12T10:02:00.000Z', last_seen_at: '2026-06-12T10:03:00.000Z', completed_at: '2026-06-12T10:03:00.000Z', turn_count: 1 },
|
|
470
|
+
},
|
|
471
|
+
},
|
|
472
|
+
},
|
|
473
|
+
}, null, 2));
|
|
474
|
+
const state = {
|
|
475
|
+
current_phase: 'ralplan',
|
|
476
|
+
handoff_artifacts: {
|
|
477
|
+
ralplan_consensus_gate: {
|
|
478
|
+
complete: true,
|
|
479
|
+
sequence: ['architect-review', 'critic-review'],
|
|
480
|
+
ralplan_architect_review: {
|
|
481
|
+
agent_role: 'architect',
|
|
482
|
+
provenance_kind: 'native_subagent',
|
|
483
|
+
verdict: 'approve',
|
|
484
|
+
session_id: sessionId,
|
|
485
|
+
thread_id: 'thread-reviewer',
|
|
486
|
+
completed_at: '2026-06-12T10:02:00.000Z',
|
|
487
|
+
},
|
|
488
|
+
ralplan_critic_review: {
|
|
489
|
+
agent_role: 'critic',
|
|
490
|
+
provenance_kind: 'native_subagent',
|
|
491
|
+
verdict: 'approve',
|
|
492
|
+
session_id: sessionId,
|
|
493
|
+
thread_id: 'thread-reviewer',
|
|
494
|
+
completed_at: '2026-06-12T10:03:00.000Z',
|
|
495
|
+
},
|
|
496
|
+
},
|
|
497
|
+
},
|
|
498
|
+
};
|
|
499
|
+
const decision = canAdvanceAutopilotRalplanToUltragoal({ cwd, sessionId, currentState: state });
|
|
500
|
+
assert.equal(decision.allowed, false);
|
|
501
|
+
assert.match(buildAutopilotRalplanUltragoalGateError(decision), /distinct native subagent tracker threads/);
|
|
502
|
+
}
|
|
503
|
+
finally {
|
|
504
|
+
await rm(cwd, { recursive: true, force: true });
|
|
505
|
+
}
|
|
506
|
+
});
|
|
507
|
+
it('rejects tracker-backed native reviews composed from different sessions', async () => {
|
|
508
|
+
const cwd = await mkdtemp(join(tmpdir(), 'omx-autopilot-ralplan-cross-session-'));
|
|
509
|
+
const trackingPath = subagentTrackingPath(cwd);
|
|
510
|
+
try {
|
|
511
|
+
await mkdir(join(trackingPath, '..'), { recursive: true });
|
|
512
|
+
await writeFile(trackingPath, JSON.stringify({
|
|
513
|
+
schemaVersion: 1,
|
|
514
|
+
sessions: {
|
|
515
|
+
'sess-architect': {
|
|
516
|
+
session_id: 'sess-architect',
|
|
517
|
+
leader_thread_id: 'thread-leader-a',
|
|
518
|
+
updated_at: '2026-06-12T10:02:00.000Z',
|
|
519
|
+
threads: {
|
|
520
|
+
'thread-leader-a': { thread_id: 'thread-leader-a', kind: 'leader', first_seen_at: '2026-06-12T09:59:00.000Z', last_seen_at: '2026-06-12T09:59:00.000Z', turn_count: 1 },
|
|
521
|
+
'thread-architect': { thread_id: 'thread-architect', kind: 'subagent', first_seen_at: '2026-06-12T10:02:00.000Z', last_seen_at: '2026-06-12T10:02:00.000Z', completed_at: '2026-06-12T10:02:00.000Z', turn_count: 1 },
|
|
522
|
+
},
|
|
523
|
+
},
|
|
524
|
+
'sess-critic': {
|
|
525
|
+
session_id: 'sess-critic',
|
|
526
|
+
leader_thread_id: 'thread-leader-c',
|
|
527
|
+
updated_at: '2026-06-12T10:03:00.000Z',
|
|
528
|
+
threads: {
|
|
529
|
+
'thread-leader-c': { thread_id: 'thread-leader-c', kind: 'leader', first_seen_at: '2026-06-12T09:59:00.000Z', last_seen_at: '2026-06-12T09:59:00.000Z', turn_count: 1 },
|
|
530
|
+
'thread-critic': { thread_id: 'thread-critic', kind: 'subagent', first_seen_at: '2026-06-12T10:03:00.000Z', last_seen_at: '2026-06-12T10:03:00.000Z', completed_at: '2026-06-12T10:03:00.000Z', turn_count: 1 },
|
|
531
|
+
},
|
|
532
|
+
},
|
|
533
|
+
},
|
|
534
|
+
}, null, 2));
|
|
535
|
+
const state = {
|
|
536
|
+
current_phase: 'ralplan',
|
|
537
|
+
handoff_artifacts: {
|
|
538
|
+
ralplan_consensus_gate: {
|
|
539
|
+
complete: true,
|
|
540
|
+
sequence: ['architect-review', 'critic-review'],
|
|
541
|
+
ralplan_architect_review: {
|
|
542
|
+
agent_role: 'architect',
|
|
543
|
+
provenance_kind: 'native_subagent',
|
|
544
|
+
verdict: 'approve',
|
|
545
|
+
session_id: 'sess-architect',
|
|
546
|
+
thread_id: 'thread-architect',
|
|
547
|
+
completed_at: '2026-06-12T10:02:00.000Z',
|
|
548
|
+
},
|
|
549
|
+
ralplan_critic_review: {
|
|
550
|
+
agent_role: 'critic',
|
|
551
|
+
provenance_kind: 'native_subagent',
|
|
552
|
+
verdict: 'approve',
|
|
553
|
+
session_id: 'sess-critic',
|
|
554
|
+
thread_id: 'thread-critic',
|
|
555
|
+
completed_at: '2026-06-12T10:03:00.000Z',
|
|
556
|
+
},
|
|
557
|
+
},
|
|
558
|
+
},
|
|
559
|
+
};
|
|
560
|
+
const decision = canAdvanceAutopilotRalplanToUltragoal({ cwd, currentState: state });
|
|
561
|
+
assert.equal(decision.allowed, false);
|
|
562
|
+
assert.match(buildAutopilotRalplanUltragoalGateError(decision), /same native subagent tracker session/);
|
|
563
|
+
}
|
|
564
|
+
finally {
|
|
565
|
+
await rm(cwd, { recursive: true, force: true });
|
|
566
|
+
}
|
|
567
|
+
});
|
|
568
|
+
it('rejects stale review session ids even when transition session context exists', async () => {
|
|
569
|
+
const cwd = await mkdtemp(join(tmpdir(), 'omx-autopilot-ralplan-live-session-mismatch-'));
|
|
570
|
+
const sessionId = 'sess-autopilot-live-session';
|
|
571
|
+
const trackingPath = subagentTrackingPath(cwd);
|
|
572
|
+
try {
|
|
573
|
+
await mkdir(join(trackingPath, '..'), { recursive: true });
|
|
574
|
+
await writeFile(trackingPath, JSON.stringify({
|
|
575
|
+
schemaVersion: 1,
|
|
576
|
+
sessions: {
|
|
577
|
+
[sessionId]: {
|
|
578
|
+
session_id: sessionId,
|
|
579
|
+
leader_thread_id: 'thread-leader',
|
|
580
|
+
updated_at: '2026-06-12T10:03:00.000Z',
|
|
581
|
+
threads: {
|
|
582
|
+
'thread-leader': { thread_id: 'thread-leader', kind: 'leader', first_seen_at: '2026-06-12T09:59:00.000Z', last_seen_at: '2026-06-12T09:59:00.000Z', turn_count: 1 },
|
|
583
|
+
'thread-architect': { thread_id: 'thread-architect', kind: 'subagent', first_seen_at: '2026-06-12T10:02:00.000Z', last_seen_at: '2026-06-12T10:02:00.000Z', completed_at: '2026-06-12T10:02:00.000Z', turn_count: 1 },
|
|
584
|
+
'thread-critic': { thread_id: 'thread-critic', kind: 'subagent', first_seen_at: '2026-06-12T10:03:00.000Z', last_seen_at: '2026-06-12T10:03:00.000Z', completed_at: '2026-06-12T10:03:00.000Z', turn_count: 1 },
|
|
585
|
+
},
|
|
586
|
+
},
|
|
587
|
+
},
|
|
588
|
+
}, null, 2));
|
|
589
|
+
const state = {
|
|
590
|
+
current_phase: 'ralplan',
|
|
591
|
+
handoff_artifacts: {
|
|
592
|
+
ralplan_consensus_gate: {
|
|
593
|
+
complete: true,
|
|
594
|
+
sequence: ['architect-review', 'critic-review'],
|
|
595
|
+
ralplan_architect_review: {
|
|
596
|
+
agent_role: 'architect',
|
|
597
|
+
provenance_kind: 'native_subagent',
|
|
598
|
+
verdict: 'approve',
|
|
599
|
+
thread_id: 'thread-architect',
|
|
600
|
+
completed_at: '2026-06-12T10:02:00.000Z',
|
|
601
|
+
},
|
|
602
|
+
ralplan_critic_review: {
|
|
603
|
+
agent_role: 'critic',
|
|
604
|
+
provenance_kind: 'native_subagent',
|
|
605
|
+
verdict: 'approve',
|
|
606
|
+
session_id: 'sess-stale-critic',
|
|
607
|
+
thread_id: 'thread-critic',
|
|
608
|
+
completed_at: '2026-06-12T10:03:00.000Z',
|
|
609
|
+
},
|
|
610
|
+
},
|
|
611
|
+
},
|
|
612
|
+
};
|
|
613
|
+
const decision = canAdvanceAutopilotRalplanToUltragoal({ cwd, sessionId, currentState: state });
|
|
614
|
+
assert.equal(decision.allowed, false);
|
|
615
|
+
assert.match(buildAutopilotRalplanUltragoalGateError(decision), /critic review session_id=sess-stale-critic does not match/);
|
|
616
|
+
}
|
|
617
|
+
finally {
|
|
618
|
+
await rm(cwd, { recursive: true, force: true });
|
|
619
|
+
}
|
|
620
|
+
});
|
|
621
|
+
it('rejects tracker-backed native reviews whose subagent threads are not completed', async () => {
|
|
622
|
+
const cwd = await mkdtemp(join(tmpdir(), 'omx-autopilot-ralplan-incomplete-thread-'));
|
|
623
|
+
const sessionId = 'sess-autopilot-incomplete-thread';
|
|
624
|
+
const trackingPath = subagentTrackingPath(cwd);
|
|
625
|
+
try {
|
|
626
|
+
await mkdir(join(trackingPath, '..'), { recursive: true });
|
|
627
|
+
await writeFile(trackingPath, JSON.stringify({
|
|
628
|
+
schemaVersion: 1,
|
|
629
|
+
sessions: {
|
|
630
|
+
[sessionId]: {
|
|
631
|
+
session_id: sessionId,
|
|
632
|
+
leader_thread_id: 'thread-leader',
|
|
633
|
+
updated_at: '2026-06-12T10:03:00.000Z',
|
|
634
|
+
threads: {
|
|
635
|
+
'thread-leader': { thread_id: 'thread-leader', kind: 'leader', first_seen_at: '2026-06-12T09:59:00.000Z', last_seen_at: '2026-06-12T09:59:00.000Z', turn_count: 1 },
|
|
636
|
+
'thread-architect': { thread_id: 'thread-architect', kind: 'subagent', first_seen_at: '2026-06-12T10:02:00.000Z', last_seen_at: '2026-06-12T10:02:00.000Z', turn_count: 1 },
|
|
637
|
+
'thread-critic': { thread_id: 'thread-critic', kind: 'subagent', first_seen_at: '2026-06-12T10:03:00.000Z', last_seen_at: '2026-06-12T10:03:00.000Z', completed_at: '2026-06-12T10:03:00.000Z', turn_count: 1 },
|
|
638
|
+
},
|
|
639
|
+
},
|
|
640
|
+
},
|
|
641
|
+
}, null, 2));
|
|
642
|
+
const state = {
|
|
643
|
+
current_phase: 'ralplan',
|
|
644
|
+
handoff_artifacts: {
|
|
645
|
+
ralplan_consensus_gate: {
|
|
646
|
+
complete: true,
|
|
647
|
+
sequence: ['architect-review', 'critic-review'],
|
|
648
|
+
ralplan_architect_review: {
|
|
649
|
+
agent_role: 'architect',
|
|
650
|
+
provenance_kind: 'native_subagent',
|
|
651
|
+
verdict: 'approve',
|
|
652
|
+
session_id: sessionId,
|
|
653
|
+
thread_id: 'thread-architect',
|
|
654
|
+
completed_at: '2026-06-12T10:02:00.000Z',
|
|
655
|
+
},
|
|
656
|
+
ralplan_critic_review: {
|
|
657
|
+
agent_role: 'critic',
|
|
658
|
+
provenance_kind: 'native_subagent',
|
|
659
|
+
verdict: 'approve',
|
|
660
|
+
session_id: sessionId,
|
|
661
|
+
thread_id: 'thread-critic',
|
|
662
|
+
completed_at: '2026-06-12T10:03:00.000Z',
|
|
663
|
+
},
|
|
664
|
+
},
|
|
665
|
+
},
|
|
666
|
+
};
|
|
667
|
+
const decision = canAdvanceAutopilotRalplanToUltragoal({ cwd, sessionId, currentState: state });
|
|
668
|
+
assert.equal(decision.allowed, false);
|
|
669
|
+
assert.match(buildAutopilotRalplanUltragoalGateError(decision), /architect tracker thread thread-architect is not completed/);
|
|
670
|
+
}
|
|
671
|
+
finally {
|
|
672
|
+
await rm(cwd, { recursive: true, force: true });
|
|
673
|
+
}
|
|
674
|
+
});
|
|
9
675
|
it('rejects native review evidence from the session leader even when malformed tracking marks it as subagent', async () => {
|
|
10
676
|
const cwd = await mkdtemp(join(tmpdir(), 'omx-autopilot-ralplan-leader-spoof-'));
|
|
11
677
|
const sessionId = 'sess-autopilot-leader-spoof';
|
|
@@ -98,6 +764,7 @@ describe('autopilot ralplan gate', () => {
|
|
|
98
764
|
kind: 'subagent',
|
|
99
765
|
first_seen_at: '2026-05-28T18:34:51.000Z',
|
|
100
766
|
last_seen_at: '2026-05-28T18:34:51.000Z',
|
|
767
|
+
completed_at: '2026-05-28T18:34:51.000Z',
|
|
101
768
|
turn_count: 1,
|
|
102
769
|
mode: 'architect',
|
|
103
770
|
},
|
|
@@ -106,6 +773,7 @@ describe('autopilot ralplan gate', () => {
|
|
|
106
773
|
kind: 'subagent',
|
|
107
774
|
first_seen_at: '2026-05-28T18:35:10.000Z',
|
|
108
775
|
last_seen_at: '2026-05-28T18:35:10.000Z',
|
|
776
|
+
completed_at: '2026-05-28T18:35:10.000Z',
|
|
109
777
|
turn_count: 1,
|
|
110
778
|
mode: 'critic',
|
|
111
779
|
},
|