@c4a/context-cli 0.6.17 → 0.6.18

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@c4a/context-cli",
-  "version": "0.6.17",
+  "version": "0.6.18",
   "type": "module",
   "description": "Local runtime and Agent integration for traceable knowledge production",
   "license": "MIT",
@@ -24,7 +24,7 @@
   },
   "dependencies": {
     "@c4a/agent-graph": "0.2.6",
-    "@c4a/context": "0.6.17",
+    "@c4a/context": "0.6.18",
     "commander": "^11.0.0",
     "fast-xml-parser": "^5.10.1",
     "handlebars": "^4.7.8",

package/plugins/VERSION

@@ -1 +1 @@
-0.6.17
+0.6.18

package/plugins/claude/.claude-plugin/plugin.json

@@ -1,7 +1,7 @@
 {
   "name": "c4a",
   "description": "Start or continue a project-local knowledge workspace through one graph-routed entry.",
-  "version": "0.6.17",
+  "version": "0.6.18",
   "author": {
     "name": "c4a"
   },

package/plugins/claude/commands/context.md

@@ -139,6 +139,11 @@ Treat `workflow.current` as the current-step authority:
    `inspection_action` resource only while inspecting the decision, and read a
    `resolution_action` resource only after the user confirms the Gate. Neither
    replaces ordinary required resources.
+   For ordinary Knowledge Review, follow the Route-selected dialogue: do not
+   advertise force approval in the initial prompt. If the user cannot access
+   the report, only their exact current-conversation reply `强制批准` authorizes
+   the returned force-approval resolution command; generic approval wording
+   does not.
 3. Execute only `commands` returned by the Route, preserving revision and
    authority flags exactly. A command marked `after-human-confirmation` waits
    for the current Gate decision. Run a command whose `execution.target` is `agent-host`
@@ -146,6 +151,9 @@ Treat `workflow.current` as the current-step authority:
    not inside a restricted child sandbox. Follow the Route-selected procedure
    for its audit and approval contract; never invent a payload, destination, or
    substitute command.
+   A Context command is complete only after its process returns an exit code and
+   receipt. If the host reports that it is still running, keep polling that same
+   invocation; never start a second Context write command in parallel.
    Treat a code-extraction batch preview as one Route action. Read its complete
    index-unit report and keep same-kind capability or scale decisions in the
    single returned Gate; do not ask about modules one by one. A non-delegatable

package/plugins/codex/.codex-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "c4a",
-  "version": "0.6.17",
+  "version": "0.6.18",
   "description": "Start or continue a project-local knowledge workspace through one graph-routed entry.",
   "author": {
     "name": "c4a"
@@ -18,7 +18,7 @@
   "skills": "./skills/",
   "interface": {
     "displayName": "C4A Context",
-    "shortDescription": "Initialize and advance a local, source-linked project knowledge workspace.\nv0.6.17",
+    "shortDescription": "Initialize and advance a local, source-linked project knowledge workspace.\nv0.6.18",
     "longDescription": "Create a Context workspace and use agent-guided next steps to register sources, run extraction, review candidates, build package outputs, and verify health without silently mutating source repositories.",
     "developerName": "c4a",
     "category": "Productivity",

package/plugins/codex/skills/context/SKILL.md

@@ -151,6 +151,11 @@ Treat `workflow.current` as the current-step authority:
    `inspection_action` resource only while inspecting the decision, and read a
    `resolution_action` resource only after the user confirms the Gate. Neither
    replaces ordinary required resources.
+   For ordinary Knowledge Review, follow the Route-selected dialogue: do not
+   advertise force approval in the initial prompt. If the user cannot access
+   the report, only their exact current-conversation reply `强制批准` authorizes
+   the returned force-approval resolution command; generic approval wording
+   does not.
 3. Execute only `commands` returned by the Route, preserving revision and
    authority flags exactly. A command marked `after-human-confirmation` waits
    for the current Gate decision. Run a command whose `execution.target` is `agent-host`
@@ -158,6 +163,9 @@ Treat `workflow.current` as the current-step authority:
    not inside a restricted child sandbox. Follow the Route-selected procedure
    for its audit and approval contract; never invent a payload, destination, or
    substitute command.
+   A Context command is complete only after its process returns an exit code and
+   receipt. If the host reports that it is still running, keep polling that same
+   invocation; never start a second Context write command in parallel.
    Treat a code-extraction batch preview as one Route action. Read its complete
    index-unit report and keep same-kind capability or scale decisions in the
    single returned Gate; do not ask about modules one by one. A non-delegatable

package/plugins/cursor/.cursor-plugin/plugin.json

@@ -1,7 +1,7 @@
 {
   "name": "c4a",
   "displayName": "C4A Context",
-  "version": "0.6.17",
+  "version": "0.6.18",
   "description": "Start or continue a project-local knowledge workspace through one graph-routed entry.",
   "author": {
     "name": "Context4AI",

package/plugins/cursor/commands/c4a-context.md

@@ -143,6 +143,11 @@ Treat `workflow.current` as the current-step authority:
    `inspection_action` resource only while inspecting the decision, and read a
    `resolution_action` resource only after the user confirms the Gate. Neither
    replaces ordinary required resources.
+   For ordinary Knowledge Review, follow the Route-selected dialogue: do not
+   advertise force approval in the initial prompt. If the user cannot access
+   the report, only their exact current-conversation reply `强制批准` authorizes
+   the returned force-approval resolution command; generic approval wording
+   does not.
 3. Execute only `commands` returned by the Route, preserving revision and
    authority flags exactly. A command marked `after-human-confirmation` waits
    for the current Gate decision. Run a command whose `execution.target` is `agent-host`
@@ -150,6 +155,9 @@ Treat `workflow.current` as the current-step authority:
    not inside a restricted child sandbox. Follow the Route-selected procedure
    for its audit and approval contract; never invent a payload, destination, or
    substitute command.
+   A Context command is complete only after its process returns an exit code and
+   receipt. If the host reports that it is still running, keep polling that same
+   invocation; never start a second Context write command in parallel.
    Treat a code-extraction batch preview as one Route action. Read its complete
    index-unit report and keep same-kind capability or scale decisions in the
    single returned Gate; do not ask about modules one by one. A non-delegatable

package/plugins/skills/c4a-context/SKILL.md

@@ -151,6 +151,11 @@ Treat `workflow.current` as the current-step authority:
    `inspection_action` resource only while inspecting the decision, and read a
    `resolution_action` resource only after the user confirms the Gate. Neither
    replaces ordinary required resources.
+   For ordinary Knowledge Review, follow the Route-selected dialogue: do not
+   advertise force approval in the initial prompt. If the user cannot access
+   the report, only their exact current-conversation reply `强制批准` authorizes
+   the returned force-approval resolution command; generic approval wording
+   does not.
 3. Execute only `commands` returned by the Route, preserving revision and
    authority flags exactly. A command marked `after-human-confirmation` waits
    for the current Gate decision. Run a command whose `execution.target` is `agent-host`
@@ -158,6 +163,9 @@ Treat `workflow.current` as the current-step authority:
    not inside a restricted child sandbox. Follow the Route-selected procedure
    for its audit and approval contract; never invent a payload, destination, or
    substitute command.
+   A Context command is complete only after its process returns an exit code and
+   receipt. If the host reports that it is still running, keep polling that same
+   invocation; never start a second Context write command in parallel.
    Treat a code-extraction batch preview as one Route action. Read its complete
    index-unit report and keep same-kind capability or scale decisions in the
    single returned Gate; do not ask about modules one by one. A non-delegatable

package/providers/context/actions/apply-forced-review.yaml

@@ -0,0 +1,5 @@
+schema: agent-graph.action.v1
+id: apply-forced-review
+runner: host
+effect: write
+handler: context.review.force-approve-current

package/providers/context/actions/revise-code-index-audit.yaml

@@ -0,0 +1,5 @@
+schema: agent-graph.action.v1
+id: revise-code-index-audit
+runner: host
+effect: write
+handler: context.project.revise-code-index

package/providers/context/actions/submit-code-index-audit.yaml

@@ -0,0 +1,6 @@
+schema: agent-graph.action.v1
+id: submit-code-index-audit
+runner: host
+effect: write
+handler: context.code-index-audit.submit
+inputSchema: schemas/code-index-audit-decision.schema.json

package/providers/context/codes.yaml

@@ -29,6 +29,9 @@ codes:
   - { code: route.extract.ownership-required, kind: route-reason, summary: Revise the index plan so every projected page has one output owner., document: resources/procedures/code-extraction.md }
   - { code: route.extract.scale-limit-exceeded, kind: route-reason, summary: At least one index unit exceeds the supported page limit., document: resources/procedures/code-extraction.md }
   - { code: route.extract.pending-target, kind: route-reason, summary: Extract the next pending code target before another workflow family., document: resources/procedures/code-extraction.md }
+  - { code: route.extract.audit-input-required, kind: route-reason, summary: The code-index audit requires specific source or protocol material that is not currently available., document: resources/procedures/code-index-audit.md }
+  - { code: route.extract.audit-revision-required, kind: route-reason, summary: Revise the code-index plan and content for issues confirmed by the Agent audit., document: resources/procedures/code-index-audit.md }
+  - { code: route.extract.audit-required, kind: route-reason, summary: The complete proposed code index requires one Agent semantic audit decision., document: resources/procedures/code-index-audit.md }
   - { code: route.document.classification-required, kind: route-reason, summary: A captured document needs an evidence-backed collection decision., document: resources/procedures/document-classification.md }
   - { code: route.prose.configuration-required, kind: route-reason, summary: A document target is missing a complete prose lifecycle declaration., document: resources/procedures/project-configuration.md }
   - { code: route.structure.pending-target, kind: route-reason, summary: Prepare the next document structure slot from source evidence., document: resources/procedures/prose-align.md }

package/providers/context/graphs/workspace.yaml

@@ -172,6 +172,7 @@ nodes:
     kind: gate
     reasonCode: route.review.decision-required
     inspectionAction: actions/inspect-review.yaml
+    resolutionAction: actions/apply-forced-review.yaml
     delegated:
       inspection: skip
       resolutionAction: actions/apply-managed-review.yaml
@@ -440,6 +441,51 @@ nodes:
         - resources/procedures/code-extraction.md
         - resources/views/source-boundary.yaml
 
+  - id: resolve-code-index-audit-input
+    kind: gate
+    reasonCode: route.extract.audit-input-required
+    resolutionAction: actions/submit-code-index-audit.yaml
+    gate:
+      id: code-index-audit-input
+      prompt: Provide the specific missing source or protocol material required to complete the code-index audit.
+      delegatable: false
+    satisfiedBy:
+      - path: extract.audit_input_required
+        equals: false
+    resources:
+      required:
+        - resources/procedures/code-index-audit.md
+        - resources/dialogue/human-gates.md
+        - resources/views/code-index-audit.yaml
+
+  - id: revise-code-index-audit
+    kind: action
+    reasonCode: route.extract.audit-revision-required
+    action: actions/revise-code-index-audit.yaml
+    satisfiedBy:
+      - path: extract.audit_revision_required
+        equals: false
+    resources:
+      required:
+        - resources/procedures/code-index-audit.md
+        - resources/semantic/code-index/classification.md
+        - resources/views/code-index-audit.yaml
+      recommended:
+        - resources/manuals/reference/code-extractors.md
+
+  - id: audit-code-index
+    kind: action
+    reasonCode: route.extract.audit-required
+    action: actions/submit-code-index-audit.yaml
+    satisfiedBy:
+      - path: extract.audit_resolved
+        equals: true
+    resources:
+      required:
+        - resources/procedures/code-index-audit.md
+        - resources/semantic/code-index/classification.md
+        - resources/views/code-index-audit.yaml
+
   - id: classify-document
     kind: gate
     reasonCode: route.document.classification-required
@@ -501,6 +547,7 @@ nodes:
     kind: gate
     reasonCode: route.review.decision-required
     inspectionAction: actions/inspect-review.yaml
+    resolutionAction: actions/apply-forced-review.yaml
     delegated:
       inspection: skip
       resolutionAction: actions/apply-managed-review.yaml
@@ -689,7 +736,10 @@ edges:
   - { from: preview-extraction-batch, to: repair-extraction-ownership }
   - { from: repair-extraction-ownership, to: revise-extraction-scale }
   - { from: revise-extraction-scale, to: extract-next }
-  - { from: extract-next, to: review-current-batch }
+  - { from: extract-next, to: resolve-code-index-audit-input }
+  - { from: resolve-code-index-audit-input, to: revise-code-index-audit, kind: gatedBy }
+  - { from: revise-code-index-audit, to: audit-code-index }
+  - { from: audit-code-index, to: review-current-batch }
   - { from: review-current-batch, to: apply-managed-review, kind: gatedBy }
   - { from: apply-managed-review, to: classify-document }
   - { from: classify-document, to: configure-prose-lifecycle, kind: gatedBy }

package/providers/context/manifest.json

@@ -2,14 +2,14 @@
   "schema": "agent-graph.bundle.v1",
   "provider": {
     "id": "c4a/context",
-    "version": "0.6.17"
+    "version": "0.6.18"
   },
   "providerManifest": "provider.yaml",
   "graphs": [
     {
       "id": "workspace",
       "path": "graphs/workspace.yaml",
-      "digest": "sha256:29a79bec9dd37dd9973fb43ab52090c90e4ba8cf734f85729a56ac1f0ec22f91"
+      "digest": "sha256:010fae5620a92e492de41d1213c34ad4013681ff2f72cc4f8e30d29ad4359881"
     }
   ],
   "actions": [
@@ -23,6 +23,11 @@
       "path": "actions/align-next.yaml",
       "digest": "sha256:8f2a7203d5b631c0b1e6cc14d4d70acbf8ac0197b3cfd19fa3111212ffb47498"
     },
+    {
+      "id": "apply-forced-review",
+      "path": "actions/apply-forced-review.yaml",
+      "digest": "sha256:ed530819831b68824bd373dec399c73a30d7eeebd2ef72ae0eab3e7a60cad80f"
+    },
     {
       "id": "apply-managed-review",
       "path": "actions/apply-managed-review.yaml",
@@ -158,13 +163,28 @@
       "path": "actions/restore-repository-sources.yaml",
       "digest": "sha256:8e44f0dc15ac9b0eea8af62dcefd315d7c03a52dab1cda373537d275ae7383dc"
     },
+    {
+      "id": "revise-code-index-audit",
+      "path": "actions/revise-code-index-audit.yaml",
+      "digest": "sha256:0ea9c41b2b698bf3a779381b42d6cd9b7eb9d54718498f3b2875eab4c23a76f6"
+    },
     {
       "id": "revise-document",
       "path": "actions/revise-document.yaml",
       "digest": "sha256:605d3808670582da69881c9a2eb8b0de21b42fbdd04c1d04a8fa93513989a08e"
+    },
+    {
+      "id": "submit-code-index-audit",
+      "path": "actions/submit-code-index-audit.yaml",
+      "digest": "sha256:30ecc7cf9fcae232f9d3075d4410f5753c599bee148e06bf08afe4c219bb45c4"
     }
   ],
   "resources": [
+    {
+      "id": "context.code-index-audit",
+      "path": "resources/views/code-index-audit.yaml",
+      "digest": "sha256:b74c2457d2cc2cb37183b6dd1da4c14f198baec56a6cbf6b78c0a1013b38cde6"
+    },
     {
       "id": "context.document-optimization-current",
       "path": "resources/views/document-optimization-current.yaml",
@@ -203,7 +223,7 @@
     {
       "id": "context.sdk.project-api",
       "path": "resources/manuals/reference/project-api.md",
-      "digest": "sha256:27c0f9523fa747c9acf7438f8e05993a83ea9942e962bfa0dfb879c9f927cc28"
+      "digest": "sha256:5fe640181630d7a363aec785ff8c76963bd41f797a4546d903fd6d5c1e7fd82b"
     },
     {
       "id": "context.sdk.template-variables",
@@ -303,7 +323,7 @@
     {
       "id": "dialogue.knowledge-review",
       "path": "resources/dialogue/knowledge-review.md",
-      "digest": "sha256:be908d5c2092790a2b3b24008367285a896c76e705685358549492a4f2eefa6b"
+      "digest": "sha256:26699582f0d9c014e84e4b21b5cf591614edbe16b544d6145c8dd61112bc1339"
     },
     {
       "id": "dialogue.package-output",
@@ -343,12 +363,17 @@
     {
       "id": "procedure.close-and-build",
       "path": "resources/procedures/close-and-build.md",
-      "digest": "sha256:bc7cc7aea9fd6796e71cc50c8ec44446fc7c576c9ddc3f6fe080d25488838592"
+      "digest": "sha256:4f68373f9406dee8607a7b814e97dcd25c8508c2113f0f50152af1ef24ea33ef"
     },
     {
       "id": "procedure.code-extraction",
       "path": "resources/procedures/code-extraction.md",
-      "digest": "sha256:b32901a9b4e68f4b8518e3bc6dfa00a010cc0a14ee2dbdf2b454948791c8a058"
+      "digest": "sha256:1e5bbf399bdf442c28e52f5f8faeb16e8f5f8fdd94f6ccdfae87ec63ded478b7"
+    },
+    {
+      "id": "procedure.code-index-audit",
+      "path": "resources/procedures/code-index-audit.md",
+      "digest": "sha256:72f98ccb1370c58381187eba9d90808cf4c71168febf17a74147da05db13c169"
     },
     {
       "id": "procedure.document-capture",
@@ -378,7 +403,7 @@
     {
       "id": "procedure.knowledge-review",
       "path": "resources/procedures/knowledge-review.md",
-      "digest": "sha256:a5d1f85514125eee0da2fda87b9b9af1e922248dffd8e3fc98f626a8325297d2"
+      "digest": "sha256:ee67b6182b7315d60aea3ed023e0688e3a0c80d90d8ebe31f38d04708d5228d1"
     },
     {
       "id": "procedure.package-output",
@@ -507,6 +532,11 @@
     }
   ],
   "schemas": [
+    {
+      "id": "schemas/code-index-audit-decision.schema.json",
+      "path": "schemas/code-index-audit-decision.schema.json",
+      "digest": "sha256:96a1bc4b962577c8a5b000a7c619271aed8845109362a6c3e54187c74bf69919"
+    },
     {
       "id": "schemas/document-optimization-decisions.schema.json",
       "path": "schemas/document-optimization-decisions.schema.json",
@@ -542,6 +572,10 @@
       "path": "actions/align-next.yaml",
       "digest": "sha256:8f2a7203d5b631c0b1e6cc14d4d70acbf8ac0197b3cfd19fa3111212ffb47498"
     },
+    {
+      "path": "actions/apply-forced-review.yaml",
+      "digest": "sha256:ed530819831b68824bd373dec399c73a30d7eeebd2ef72ae0eab3e7a60cad80f"
+    },
     {
       "path": "actions/apply-managed-review.yaml",
       "digest": "sha256:25e2c2c215889dd13a6dec8714bc2b6357f3e01b09cc684f1e82d986e30ef064"
@@ -650,21 +684,29 @@
       "path": "actions/restore-repository-sources.yaml",
       "digest": "sha256:8e44f0dc15ac9b0eea8af62dcefd315d7c03a52dab1cda373537d275ae7383dc"
     },
+    {
+      "path": "actions/revise-code-index-audit.yaml",
+      "digest": "sha256:0ea9c41b2b698bf3a779381b42d6cd9b7eb9d54718498f3b2875eab4c23a76f6"
+    },
     {
       "path": "actions/revise-document.yaml",
       "digest": "sha256:605d3808670582da69881c9a2eb8b0de21b42fbdd04c1d04a8fa93513989a08e"
     },
+    {
+      "path": "actions/submit-code-index-audit.yaml",
+      "digest": "sha256:30ecc7cf9fcae232f9d3075d4410f5753c599bee148e06bf08afe4c219bb45c4"
+    },
     {
       "path": "codes.yaml",
-      "digest": "sha256:d2edb3ac2599e5a5acfa1ef3bd807f23dd743d4cce10a53113a511eaf314c927"
+      "digest": "sha256:387e46f7da76e3152dcd5ab61481bb3c4465624ecfbfd58994276b67b4565f5e"
     },
     {
       "path": "graphs/workspace.yaml",
-      "digest": "sha256:29a79bec9dd37dd9973fb43ab52090c90e4ba8cf734f85729a56ac1f0ec22f91"
+      "digest": "sha256:010fae5620a92e492de41d1213c34ad4013681ff2f72cc4f8e30d29ad4359881"
     },
     {
       "path": "provider.yaml",
-      "digest": "sha256:31e9ca586a2c9edab004d04be8221e46897a7b88a601e9acad603af3ba250910"
+      "digest": "sha256:d4fc8b8a219056c3f48bfa69d467e12e592a348f1c823a08ab553e8374d6af54"
     },
     {
       "path": "resources/diagnostics/projection-stale.md",
@@ -700,7 +742,7 @@
     },
     {
       "path": "resources/dialogue/knowledge-review.md",
-      "digest": "sha256:be908d5c2092790a2b3b24008367285a896c76e705685358549492a4f2eefa6b"
+      "digest": "sha256:26699582f0d9c014e84e4b21b5cf591614edbe16b544d6145c8dd61112bc1339"
     },
     {
       "path": "resources/dialogue/package-output.md",
@@ -744,7 +786,7 @@
     },
     {
       "path": "resources/manuals/reference/project-api.md",
-      "digest": "sha256:27c0f9523fa747c9acf7438f8e05993a83ea9942e962bfa0dfb879c9f927cc28"
+      "digest": "sha256:5fe640181630d7a363aec785ff8c76963bd41f797a4546d903fd6d5c1e7fd82b"
     },
     {
       "path": "resources/manuals/reference/template-variables.md",
@@ -752,11 +794,15 @@
     },
     {
       "path": "resources/procedures/close-and-build.md",
-      "digest": "sha256:bc7cc7aea9fd6796e71cc50c8ec44446fc7c576c9ddc3f6fe080d25488838592"
+      "digest": "sha256:4f68373f9406dee8607a7b814e97dcd25c8508c2113f0f50152af1ef24ea33ef"
     },
     {
       "path": "resources/procedures/code-extraction.md",
-      "digest": "sha256:b32901a9b4e68f4b8518e3bc6dfa00a010cc0a14ee2dbdf2b454948791c8a058"
+      "digest": "sha256:1e5bbf399bdf442c28e52f5f8faeb16e8f5f8fdd94f6ccdfae87ec63ded478b7"
+    },
+    {
+      "path": "resources/procedures/code-index-audit.md",
+      "digest": "sha256:72f98ccb1370c58381187eba9d90808cf4c71168febf17a74147da05db13c169"
     },
     {
       "path": "resources/procedures/document-capture.md",
@@ -780,7 +826,7 @@
     },
     {
       "path": "resources/procedures/knowledge-review.md",
-      "digest": "sha256:a5d1f85514125eee0da2fda87b9b9af1e922248dffd8e3fc98f626a8325297d2"
+      "digest": "sha256:ee67b6182b7315d60aea3ed023e0688e3a0c80d90d8ebe31f38d04708d5228d1"
     },
     {
       "path": "resources/procedures/package-output.md",
@@ -902,6 +948,10 @@
       "path": "resources/semantic/compile/index.md",
       "digest": "sha256:4145f6bdc2855e354d00d311448584c3ed4b73849fc1277a2ed72d7be84e9e7c"
     },
+    {
+      "path": "resources/views/code-index-audit.yaml",
+      "digest": "sha256:b74c2457d2cc2cb37183b6dd1da4c14f198baec56a6cbf6b78c0a1013b38cde6"
+    },
     {
       "path": "resources/views/document-optimization-current.yaml",
       "digest": "sha256:5c98530b9a33e09445ae5e0f18151f66975106b6a0d327ac27875497c2f7f76d"
@@ -938,6 +988,10 @@
       "path": "resources/views/workspace-current.yaml",
       "digest": "sha256:712d146a1538e47d09c0cfad9961c5ad1c7ebb7e9796a52c0b0bdb6007f50dc1"
     },
+    {
+      "path": "schemas/code-index-audit-decision.schema.json",
+      "digest": "sha256:96a1bc4b962577c8a5b000a7c619271aed8845109362a6c3e54187c74bf69919"
+    },
     {
       "path": "schemas/document-optimization-decisions.schema.json",
       "digest": "sha256:a21e84dbdcc2bad85970ebc3171f58176b480f751076745424c7f9742b972f5c"
@@ -966,5 +1020,5 @@
   "graphDependencies": {
     "workspace": []
   },
-  "digest": "sha256:5da3c1272fa593c6eb36bbb33acc303e01dae0788f5f6ff5389835a76f8bd2df"
+  "digest": "sha256:a3f1974e2359466c24d0853d3326ab7799d5e3b02cb611c667be726f09843ee3"
 }

package/providers/context/provider.yaml

@@ -1,6 +1,6 @@
 schema: agent-graph.provider.v1
 id: c4a/context
-version: 0.6.17
+version: 0.6.18
 name: Context workflow
 description: Internal work contract for Context knowledge workspaces.
 graphs:

package/providers/context/resources/dialogue/knowledge-review.md

@@ -18,6 +18,15 @@ The user does not need to create a payload file; the Agent may write the pasted
 payload to ignored scratch storage for the CLI command. Never derive a payload
 from HTML, candidate ids, snapshots, or a default decision.
 
+Do not mention force approval when first presenting Review. If the user replies
+without a Payload that they approve or want to continue, explain that the
+report Payload remains the normal path. Only at that point, when the report is
+unavailable to them, tell them they may explicitly reply with the exact phrase
+`强制批准` to approve the complete current scope without per-candidate choices.
+Execute the Route's `after-human-confirmation` force-approval command only after
+that exact phrase appears in the current conversation. Phrases such as `我批准`,
+`继续`, or `全部通过` do not invoke this escape path.
+
 Do not open Review for one page or module while another confirmed item in the
 same round is still being generated. If a repeat codegraph run has no delta,
 state that existing approvals were preserved and no Review gate remains.

package/providers/context/resources/manuals/reference/project-api.md

@@ -951,11 +951,18 @@ review html --all --open --format json`, confirm the returned `opened` field,
 and wait for the user-copied payload. They should not run the phase as an
 automatic approval step or synthesize a payload themselves.
 
-The only batch-wide exception is explicit current-conversation fully managed
-authority. In that mode, follow the `context status --managed` route to
+One batch-wide path is explicit current-conversation fully managed authority.
+In that mode, follow the `context status --managed` route to
 `context review approve-all ... --managed`; the CLI validates the exact current
 scope before applying one default-approved decision.
 
+Ordinary mode has a separate user-confirmed escape path for environments where
+the Review report cannot be opened. Do not advertise it in the initial Review
+prompt. After the user reports that limitation, the exact conversation phrase
+`强制批准` authorizes only the current Route's revision-bound
+`context review approve-all ... --force` command. Other generic approval or
+continue wording does not invoke it.
+
 The gate is batch-scoped: prose waits for every planned View across all active
 structure slots and every declared `pendingStructureTargets` item in the round;
 codegraph waits for every pending extract phase in the confirmed module round.

package/providers/context/resources/procedures/close-and-build.md

@@ -26,3 +26,11 @@ package inventory. Revision sidecars are never emitted as separate knowledge.
 Run only the current route command, then evaluate again. A successful build
 means the currently declared scope is current; newly captured or newly declared
 targets can reopen earlier graph nodes.
+# Code-index audit record
+
+When a package selects approved `codegraph` pages, build requires the current
+batch-level Agent audit to be accepted. The complete report remains in the
+Context workspace at `knowledge/code-index-audit.json` and is not copied into
+`dist/` or uploaded with the package. `context-build-inventory.json` stores only
+the report digest, decision, and compact package summary needed to identify the
+audited build.

package/providers/context/resources/procedures/code-extraction.md

@@ -144,3 +144,17 @@ extractor, keep the lifecycle in `extractCustom()` and implement only the
 missing project adapter. If reliable syntax facts cannot be produced, stop at
 configuration and report the generic capability gap rather than emitting an
 empty or guessed graph.
+# Agent audit after extraction
+
+After every complete code-extraction batch, the Route produces one batch-level
+code-index audit. Mechanical signals cover content depth, evidence scope,
+declared-source coverage, and structured handoffs. They are review evidence,
+not an automatic numeric rejection: the Agent must inspect the affected pages
+and submit one `accept`, `revise`, or `request-input` decision for all index
+units together.
+
+Do not split this into one confirmation per module. In fully managed operation,
+real issues select `revise` and the Route returns through project configuration,
+batch Preview, extraction, and a new audit until the index is acceptable. A
+false positive may be accepted only with a concrete inspected reason. Ask the
+user only when reliable correction needs unavailable material or access.

package/providers/context/resources/procedures/code-index-audit.md

@@ -0,0 +1,43 @@
+---
+id: procedure.code-index-audit
+kind: procedure
+mediaType: text/markdown
+---
+
+# Code-index Agent audit
+
+This Route is a required semantic review of the complete proposed code index.
+The CLI reports mechanical signals; the Agent reads the affected candidate
+content and evidence, compares registered sources with the user-confirmed
+scope, and decides whether the index is fit for its declared output profiles.
+
+The report is not a numeric rejection. A signal may be a false positive, but
+it may not be ignored. Every elevated signal must receive an assessment tied
+to inspected content. Submit one `context.code-index-audit-decision.v1`
+payload for the complete batch.
+
+## Decisions
+
+- `accept`: the requested scope is represented and every real content-depth,
+  evidence-scope, and relationship issue is resolved. An acceptable or
+  non-applicable signal needs a concrete reason.
+- `revise`: one or more real problems remain. Identify the affected units and
+  describe changes to scope, aggregation, sections, evidence, or structured
+  handoffs. The Route returns to `src/index.ts`, Preview, extraction, and a new
+  audit revision.
+- `request-input`: reliable revision requires material unavailable in the
+  registered sources, such as an external protocol or missing source boundary.
+  Ask only for that material, then submit a new decision.
+
+## Fully managed operation
+
+Fully managed authority does not auto-accept this audit. When a signal is a
+real problem, choose `revise` and continue the revision loop without asking the
+user. Repeat until the proposed index matches the requested scope and its
+module maps, contracts, and handoffs contain useful source-backed explanation.
+Pause only for unavailable material, source access, tool failure, or a
+concrete no-progress blocker.
+
+After each decision, briefly tell the user which decision was selected, the
+affected units, and the next Graph path. Do not stop a managed run merely to
+announce a successful automatic revision.

package/providers/context/resources/procedures/knowledge-review.md

@@ -26,6 +26,12 @@ Without explicit session-managed authority:
 - let the user approve or reject candidates;
 - apply the exact returned decision payload.
 
+The ordinary Route also carries a revision-bound force-approval resolution
+Action as an escape path. Do not advertise it when first presenting Review.
+Use it only after the user cannot use the report and explicitly replies with
+the exact phrase `强制批准` in the current conversation. It approves the complete
+current scope atomically; no candidate-specific decisions are inferred.
+
 With explicit session-managed authority, Context may approve the complete
 current batch atomically. The authority exists only in the current
 conversation. It does not bypass source permission, validation, close, or

package/providers/context/resources/views/code-index-audit.yaml

@@ -0,0 +1,6 @@
+schema: agent-graph.resource.v1
+id: context.code-index-audit
+kind: context-view
+mediaType: text/markdown
+description: Current code-index content-depth, evidence-scope, relationship, source-coverage signals, and prior Agent decisions.
+materializer: actions/materialize-context-view.yaml