deadbyte-mcp 0.12.0 → 0.14.0
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/.gitattributes +7 -0
- package/AGENTS.md +25 -0
- package/CONTEXT.md +3 -3
- package/MANIFEST.SHA256 +89 -45
- package/R41-PARENT.json +15 -0
- package/R42-PARENT.json +19 -0
- package/README.txt +2 -2
- package/bin/appcontainer-stage.obj +0 -0
- package/bin/bootstrap-advapi32.obj +0 -0
- package/bin/bootstrap-exitcode.obj +0 -0
- package/bin/bootstrap-kernel32.obj +0 -0
- package/bin/child-control-probe.obj +0 -0
- package/bin/child-control-stage.obj +0 -0
- package/bin/contained-reverse-worker.obj +0 -0
- package/bin/contained-transform-worker.obj +0 -0
- package/bin/containment-probe.obj +0 -0
- package/bin/deadbyte-contain.obj +0 -0
- package/bin/deadbyte-exec.obj +0 -0
- package/bin/deadbyte-process-host.obj +0 -0
- package/bin/deadbyte-tunnel-host.obj +0 -0
- package/controller/README.TXT +1 -1
- package/controller/deadbyte-controller.ps1 +1 -1
- package/controller/deadbyte-desktop-policy.json +5 -5
- package/controller/deadbyte-process-policy.json +4 -4
- package/docs/ARCHITECTURE.md +1 -1
- package/docs/superpowers/specs/2026-09-20-r41-deterministic-memory-design.md +132 -0
- package/harness/r42-agent-evals.json +48 -0
- package/harness/task-contract.schema.json +33 -0
- package/package.json +11 -7
- package/proof/CONTAINMENT-BUILD.txt +6 -6
- package/scripts/agent-eval-r42.mjs +46 -0
- package/scripts/build-containment.ps1 +24 -14
- package/scripts/build-desktop-r40.ps1 +15 -9
- package/scripts/deferred-slot-operation-r41.mjs +127 -0
- package/scripts/deferred-slot-operation-r42.mjs +126 -0
- package/scripts/final-closure-r41.mjs +116 -0
- package/scripts/final-closure-r42.mjs +116 -0
- package/scripts/final-closure-verify-r41.mjs +215 -0
- package/scripts/final-closure-verify-r42.mjs +223 -0
- package/scripts/gate-windows-r41.ps1 +48 -0
- package/scripts/gate-windows-r42.ps1 +54 -0
- package/scripts/init-desktop-policy-r40.ps1 +10 -4
- package/scripts/release-parity-r41.mjs +221 -0
- package/scripts/release-parity-r42.mjs +235 -0
- package/scripts/release-parity-tests-r41.ps1 +9 -0
- package/scripts/release-parity-tests-r42.ps1 +6 -0
- package/scripts/verify-memory-r41.mjs +88 -0
- package/scripts/verify-r41-parent.mjs +61 -0
- package/scripts/verify-r42-parent.mjs +101 -0
- package/scripts/windows-gate-evidence-r41.mjs +53 -0
- package/scripts/windows-gate-evidence-r42.mjs +64 -0
- package/src/agent-eval-r42.mjs +73 -0
- package/src/agent-eval-runner-r42.mjs +138 -0
- package/src/agent-governance-r42.mjs +142 -0
- package/src/autonomous-context-engine.mjs +4 -1
- package/src/autonomous-loop-state.mjs +10 -0
- package/src/autonomous-mcp-tools.mjs +21 -6
- package/src/autonomous-output-contracts.mjs +18 -2
- package/src/autonomous-planner-contracts.mjs +3 -0
- package/src/autonomous-planner.mjs +48 -6
- package/src/autonomous-policy.mjs +10 -2
- package/src/autonomous-runtime.mjs +144 -25
- package/src/autonomous-supervisor.mjs +1 -1
- package/src/evidence-publisher-r42.mjs +76 -0
- package/src/mcp-server.mjs +15 -2
- package/src/memory-r41.mjs +692 -0
- package/src/provider-circuit.mjs +4 -4
- package/src/version.mjs +1 -1
- package/test/agent-eval-r42.test.mjs +39 -0
- package/test/agent-eval-runner-r42.test.mjs +61 -0
- package/test/agent-governance-r42.test.mjs +77 -0
- package/test/autonomous-r34-e2e.test.mjs +2 -1
- package/test/cli-entrypoint.test.mjs +1 -1
- package/test/controller-desktop-r40.test.mjs +2 -1
- package/test/core.test.mjs +1 -1
- package/test/evidence-publisher-r42.test.mjs +53 -0
- package/test/helpers/autonomous-r34-fixture.mjs +3 -3
- package/test/human-clarification-r42.test.mjs +53 -0
- package/test/memory-autonomous-integration-r41.test.mjs +31 -0
- package/test/memory-history-r41.test.mjs +126 -0
- package/test/memory-projections-r41.test.mjs +92 -0
- package/test/memory-retrieval-r41.test.mjs +124 -0
- package/test/memory-snapshot-r41.test.mjs +106 -0
- package/test/provider-responses-r42.test.mjs +90 -0
- package/test/r33-closeout-regression.test.mjs +27 -12
- package/test/r33-finalization.test.mjs +4 -4
- package/test/r41-finalization.test.mjs +38 -0
- package/test/r42-finalization.test.mjs +38 -0
- package/test/release-version.test.mjs +14 -13
- package/test/task-contract-r42.test.mjs +41 -0
|
@@ -0,0 +1,41 @@
|
|
|
1
|
+
import test from 'node:test';
|
|
2
|
+
import assert from 'node:assert/strict';
|
|
3
|
+
import { createR34Fixture } from './helpers/autonomous-r34-fixture.mjs';
|
|
4
|
+
|
|
5
|
+
const contract={
|
|
6
|
+
schema:'deadbyte.task-contract.v1',
|
|
7
|
+
scope:['src/autonomous-*','test/*r42*'],
|
|
8
|
+
non_goals:['No generic shell','No implicit grant renewal'],
|
|
9
|
+
constraints:['Preserve R41 compatibility','Use exact preconditions'],
|
|
10
|
+
evidence_requirements:['Focused tests pass','Signed journal verifies'],
|
|
11
|
+
risk_level:'high',
|
|
12
|
+
clarification_policy:'ask_when_blocked',
|
|
13
|
+
project_default:'lightweight',
|
|
14
|
+
explicit_override:'structured'
|
|
15
|
+
};
|
|
16
|
+
|
|
17
|
+
test('R42 task contract is identity-bound and visible to every planner round',async()=>{
|
|
18
|
+
const f=await createR34Fixture();
|
|
19
|
+
try {
|
|
20
|
+
const goal=await f.runtime.submit({submissionId:'r42-task-contract',rootId:'project',title:'Contract',
|
|
21
|
+
goal:'Implement bounded behavior.',acceptance:'All required evidence passes.',taskContract:contract,maxIterations:8});
|
|
22
|
+
const status=await f.runtime.status({goalId:goal.goal_id});
|
|
23
|
+
assert.equal(status.state,'pending');
|
|
24
|
+
const run=await f.runtime.run({goalId:goal.goal_id,maxCycles:1});
|
|
25
|
+
assert.equal(run.status,'input_required');
|
|
26
|
+
assert.match(run.planner_request.prompt,/deadbyte\.task-contract\.v1/);
|
|
27
|
+
assert.match(run.planner_request.prompt,/No generic shell/);
|
|
28
|
+
assert.match(run.planner_request.prompt,/Signed journal verifies/);
|
|
29
|
+
assert.match(run.planner_request.prompt,/"effective_rigor":"agentic"/);
|
|
30
|
+
} finally { await f.cleanup(); }
|
|
31
|
+
});
|
|
32
|
+
|
|
33
|
+
test('R42 task contract rejects authority widening fields and invalid risk',async()=>{
|
|
34
|
+
const f=await createR34Fixture();
|
|
35
|
+
try {
|
|
36
|
+
await assert.rejects(f.runtime.submit({submissionId:'r42-task-injected',rootId:'project',title:'Contract',
|
|
37
|
+
goal:'Reject injection.',acceptance:'Fail closed.',taskContract:{...contract,shell:'cmd.exe'}}),/unknown field|task contract/i);
|
|
38
|
+
await assert.rejects(f.runtime.submit({submissionId:'r42-task-risk',rootId:'project',title:'Contract',
|
|
39
|
+
goal:'Reject risk.',acceptance:'Fail closed.',taskContract:{...contract,risk_level:'unbounded'}}),/risk_level/i);
|
|
40
|
+
} finally { await f.cleanup(); }
|
|
41
|
+
});
|