create-agdf 0.14.3 → 0.14.4

package/README.md

@@ -17,7 +17,7 @@ If you use Codex, this is the recommended first installation:
 npx --yes @agdf/cli@latest codex
 ```
 
-Restart Codex, open a new task in the repository you want to work in, and describe the intended change. AGDF starts with the smallest permitted governance step; it does not make implementation automatic.
+Fully restart Codex, open a fresh task in the repository you want to work in, and describe the intended change. Restoring a previous task can retain stale AGDF skills. AGDF starts with the smallest permitted governance step; it does not make implementation automatic.
 
 Use a different target when Codex is not your agent surface:
 
@@ -72,6 +72,8 @@ npx --yes @agdf/cli@latest opencode-status
 npx --yes @agdf/cli@latest opencode-repo
 npx --yes @agdf/cli@latest status --surface codex
 npx --yes @agdf/cli@latest disable --surface codex --scope repository
+npx --yes @agdf/cli@latest disable --surface copilot --scope repository
+npx --yes @agdf/cli@latest disable --surface copilot --scope repository --shared
 npx --yes @agdf/cli@latest uninstall --surface codex --scope global
 npx --yes @agdf/cli@latest init
 npx --yes @agdf/cli@latest config --language en
@@ -125,6 +127,7 @@ Optional flags:
 - `--verbose` show captured host installer output and generated-file details after the concise lifecycle card
 - `--scope <repository|global>` select an explicit lifecycle mutation scope
 - `--confirm` apply a global uninstall after reviewing the default non-mutating preview
+- `--shared` use commit-capable `.github/copilot/settings.json` for an explicit shared Copilot repository disable; without it Copilot uses ignored personal-local settings
 
 If no language is provided, `create-agdf` derives the preference from the local system locale (`LC_ALL`, `LC_MESSAGES`, `LANG`, `LANGUAGE` or the Node.js runtime locale) and falls back to `en`.
 
@@ -142,7 +145,7 @@ it does not localize the CLI lifecycle card.
 - `opencode-status` reports OpenCode global config, package loadability, global native-skill completeness, installed host/plugin-SDK versions, declaration-level support for AGDF's two experimental hooks, durable repository activation, legacy compatibility and observable session signals
 - `status` reports installation, repository activation and delivery separately without mutating state
 - `runtime-checks status|enable|manual` reports the requested/effective automatic-check state or gives the exact reinstall route needed to change it
-- `disable` writes a supported repository-local opt-out while retaining global capability and durable control state
+- `disable` keeps Codex repository behavior and supports Copilot personal-local opt-out by default; Copilot shared repository effect requires `--shared`
 - `uninstall` previews and, only with `--confirm`, applies a selected global removal through supported native/owned operations
 - `opencode-repo` writes durable AGDF control configuration and templates under `.agdf/control/`; it does not copy a second OpenCode runtime surface
 - `config` writes or updates only `.agdf/control/config.json` for an already installed plugin or an existing repository
@@ -153,12 +156,41 @@ the Copilot manifest, prefixed skills, hook, required contracts and exact-versio
 is rendered from the same canonical sources. The installers atomically stage their profile under an AGDF-owned user-data marketplace, register
 that stable local path through the host CLI and verify the exposed version. Source `plugin/` therefore
 contains no generated runtime bytes and the source checkout exposes no installable root marketplace.
+
+Repository lifecycle support is deliberately asymmetric:
+
+| Surface | Personal repository opt-out | Shared repository opt-out | Repository activation | Global uninstall |
+|---|---|---|---|---|
+| Codex | supported by its existing local plugin state | no new `--shared` mode | `codex-repo` | supported |
+| Claude Code | not supported without a verified host mechanism | not supported | no new mechanism | supported |
+| GitHub Copilot | default in ignored `.github/copilot/settings.local.json` | explicit `--shared` in `.github/copilot/settings.json` | plugin discovery remains separate | supported |
+| OpenCode | not supported as disable | not supported | `opencode-repo` | supported |
+
+The personal Copilot command fails before mutation unless Git confirms the local settings path is
+ignored. It never edits `.gitignore` or `.git/info/exclude`. Existing JSON must be strict JSON;
+JSONC, comments, invalid types and symlinked paths fail closed. Only
+`enabledPlugins["agdf@agdf"]` changes. Restart Copilot and inspect `/plugin list`; inspect
+`/instructions` separately because plugin disablement does not disable `AGENTS.md`,
+`.github/copilot-instructions.md` or other applicable instructions.
 The staged plugin contains one installation-provenance marker and the shared exact-version runtime;
 routine installed validation does not depend on the GitHub
 checkout, npm cache, PATH or registry. Rerunning either command performs the explicit update and
 migrates only the exact known legacy AGDF GitHub marketplace; foreign same-name registrations fail
 closed and failed host operations restore the prior owned stage.
 
+The shared installer recognizes exact packaged snapshots for the verified AGDF-owned four-profile
+releases `0.13.6`, `0.13.7`, `0.13.8` and `0.14.1`, then rebuilds them from current canonical
+package content. Current-shape `0.14.2` and `0.14.3` stay on ordinary current validation.
+`agdf-v0.14.0` is not a release alias: its internal version is `0.13.8`, so it grants no
+compatibility. Historical lookup uses no Git or network access and does not accept unknown versions,
+partial contracts or tampered provenance. Claude's Windows-only contention recovery removes at most the one
+contained `temp_local_*` directory named by the current install command's `EPERM` rename failure and
+retries once; it does not enumerate or broadly clear the host cache.
+
+Successful installation verifies the installed version but not an already loaded session. Fully
+restart the host and start a fresh session or task. Restoring the previous session can retain stale
+AGDF skills and must not be treated as current loaded-session evidence.
+
 Copilot staging uses the independent path `<AGDF data directory>/marketplaces/agdf-copilot`. This
 prevents Copilot updates and rollbacks from replacing the shared Codex and Claude payload while the
 host-facing Marketplace identity remains `agdf` and the install identity remains `agdf@agdf`.

package/generated/.opencode/contracts/interaction.md

@@ -154,8 +154,8 @@ required decision and one neutral instruction to choose approve, request revisio
 or decline. It does not contain the exact approval value or quality outlook. It uses localized human-readable labels
 and must not contain evidence, diagnostic codes, raw control-state keys,
 allowed/forbidden inventories or machine values. The complete Run Status Card remains the operational,
-CLI and audit projection and remains available for complete detail outside the
-approval-time compact view.
+CLI and audit projection. Every ready-gate presentation renders the complete operational Run Status Card exactly once, between the compact approval-time Run Status Card and the Gate Transition Card, consumed verbatim from the code-owned status presentation; when it is undeliverable, the localized status-presentation failure line with its concrete diagnostic codes takes its place and the decision is still requested.
+The exact-approval once-only requirement applies to the compact approval-time Run Status Card and the Gate Transition Card only; the complete Run Status Card may carry the missing approval as audit data.
 
 The first scan across both cards reveals the requested decision, readiness,
 required approval, approval effect, remaining boundary and next transition.

package/generated/.opencode/skills/agdf-gate-check/SKILL.md

@@ -209,10 +209,12 @@ the primary chat card into an audit dump. Do not maintain or render a skill-loca
 
 This complete operational status table remains the status-reporting and detail
 surface. When the same response immediately requests a ready gate approval,
-render its five-field compact approval-time projection first, then the Gate
-Transition Card in the same immediately preceding assistant message, then
-invoke exactly one native question or exact-text fallback. Both cards derive
-from the same snapshot and are shown exactly once.
+render its five-field compact approval-time projection first, then the complete
+operational Run Status Card verbatim (or its localized failure line with
+diagnostic codes when undeliverable), then the Gate Transition Card in the same
+immediately preceding assistant message, then invoke exactly one native
+question or exact-text fallback. The runtime contract owns this sequence; every
+block is shown exactly once.
 
 If this skill creates or updates control artefacts, do not paste full file bodies into the chat.
 List paths, summarize the decision, name the blocker or approval needed, and keep the durable content in the files.

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

@@ -1,6 +1,6 @@
 {
   "name": "agdf",
-  "version": "0.14.3",
+  "version": "0.14.4",
   "description": "AI Governance & Delivery Framework (AGDF) is a governance framework for agentic work. Its first implementation provides a control layer for governed AI-assisted software delivery. It combines workflow skills, explicit gates, control templates and machine-readable repository checks to keep work aligned with approved scope and make evidence a prerequisite for consequential gate transitions. Supported workflows include brownfield analysis, task-plan review, QA decisions, durable run state and auditable delivery closeout. AGDF is an independent open-source project, not a security sandbox, compliance certification or autonomous release system.",
   "keywords": [
     "agentic",

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

@@ -1,6 +1,6 @@
 {
   "name": "agdf",
-  "version": "0.14.3",
+  "version": "0.14.4",
   "description": "Control layer for governed AI-assisted delivery.",
   "author": {
     "name": "Arndt Gold",

package/generated/plugins/agdf/meta/agdf-plugin.definition.json

@@ -1,7 +1,7 @@
 {
   "schemaVersion": 1,
   "id": "agdf",
-  "version": "0.14.3",
+  "version": "0.14.4",
   "displayName": "AI Governance & Delivery Framework",
   "developerName": "Arndt Gold",
   "description": "Control layer for governed AI-assisted delivery.",

package/generated/plugins/agdf/meta/contracts/interaction.md

@@ -154,8 +154,8 @@ required decision and one neutral instruction to choose approve, request revisio
 or decline. It does not contain the exact approval value or quality outlook. It uses localized human-readable labels
 and must not contain evidence, diagnostic codes, raw control-state keys,
 allowed/forbidden inventories or machine values. The complete Run Status Card remains the operational,
-CLI and audit projection and remains available for complete detail outside the
-approval-time compact view.
+CLI and audit projection. Every ready-gate presentation renders the complete operational Run Status Card exactly once, between the compact approval-time Run Status Card and the Gate Transition Card, consumed verbatim from the code-owned status presentation; when it is undeliverable, the localized status-presentation failure line with its concrete diagnostic codes takes its place and the decision is still requested.
+The exact-approval once-only requirement applies to the compact approval-time Run Status Card and the Gate Transition Card only; the complete Run Status Card may carry the missing approval as audit data.
 
 The first scan across both cards reveals the requested decision, readiness,
 required approval, approval effect, remaining boundary and next transition.

package/generated/plugins/agdf/meta/copilot-payload-baseline.json

@@ -1,7 +1,7 @@
 {
   "schema_version": 1,
   "profile_id": "copilot-runtime-plugin",
-  "max_files": 78,
-  "max_bytes": 550979,
-  "rationale": "Reviewed Copilot-only manifest, ten prefixed skills, required contracts, exact-version runtime, Delivery Path Search input-integrity provenance and evidence-plane separation. Any further increase requires review."
+  "max_files": 80,
+  "max_bytes": 568459,
+  "rationale": "Reviewed Copilot-only manifest, ten prefixed skills, required contracts, exact-version runtime, Delivery Path Search input-integrity provenance, evidence-plane separation, shared run-identity validation and the one-file distribution profile history catalogue. Measured payload is 80 files/563615 bytes; no byte-capacity increase was required. Any further increase requires review."
 }

package/generated/plugins/agdf/meta/distribution-profile-history.json

@@ -0,0 +1,125 @@
+{
+  "schema_version": 1,
+  "contracts": {
+    "five-profile-v1": {
+      "distribution_profiles": {
+        "schemaVersion": 1,
+        "marketplaceIdentities": {
+          "durable": "agdf",
+          "generatedRepository": "agdf-repo"
+        },
+        "profiles": {
+          "source-development": {
+            "runtime": "absent",
+            "installable": false,
+            "machineValidation": "unavailable"
+          },
+          "runtime-plugin": {
+            "runtime": "required",
+            "installable": true,
+            "machineValidation": "local_exact_version_digest"
+          },
+          "copilot-runtime-plugin": {
+            "runtime": "required",
+            "installable": true,
+            "machineValidation": "local_exact_version_digest_inventory"
+          },
+          "opencode-config-local": {
+            "runtime": "config_local_package",
+            "installable": true,
+            "machineValidation": "local_exact_version"
+          },
+          "portable-skills": {
+            "runtime": "absent",
+            "installable": true,
+            "machineValidation": "unavailable_or_external_required"
+          }
+        }
+      },
+      "contract_digest": "0211f4e6054a23f0a3ba302e57303adcb8f5903581e99643f803647dd156fe32"
+    },
+    "four-profile-v1": {
+      "distribution_profiles": {
+        "schemaVersion": 1,
+        "marketplaceIdentities": {
+          "durable": "agdf",
+          "generatedRepository": "agdf-repo"
+        },
+        "profiles": {
+          "source-development": {
+            "runtime": "absent",
+            "installable": false,
+            "machineValidation": "unavailable"
+          },
+          "runtime-plugin": {
+            "runtime": "required",
+            "installable": true,
+            "machineValidation": "local_exact_version_digest"
+          },
+          "opencode-config-local": {
+            "runtime": "config_local_package",
+            "installable": true,
+            "machineValidation": "local_exact_version"
+          },
+          "portable-skills": {
+            "runtime": "absent",
+            "installable": true,
+            "machineValidation": "unavailable_or_external_required"
+          }
+        }
+      },
+      "contract_digest": "9987baa4ea91e28a096cac2be8daa876f1c0593e7edbfe2e96b713814aa65ce8"
+    }
+  },
+  "releases": {
+    "0.13.6": {
+      "contract_id": "four-profile-v1",
+      "provenance_schema_version": 1,
+      "profile_id": "runtime-plugin",
+      "entry_digest": "1c21c3fa99cdd788994013f25cbe606ffdcea2cc91d53c679cf061567f652bf3",
+      "status": "supported"
+    },
+    "0.13.7": {
+      "contract_id": "four-profile-v1",
+      "provenance_schema_version": 1,
+      "profile_id": "runtime-plugin",
+      "entry_digest": "6d5383537d2640dc13773f2929dd94d0e7e219e82d08418464c77786b475c3ee",
+      "status": "supported"
+    },
+    "0.13.8": {
+      "contract_id": "four-profile-v1",
+      "provenance_schema_version": 1,
+      "profile_id": "runtime-plugin",
+      "entry_digest": "405e8cfa72ec2e469354dab52197675783b3723adb969a0692e56ba7c95c85c8",
+      "status": "supported"
+    },
+    "0.14.1": {
+      "contract_id": "four-profile-v1",
+      "provenance_schema_version": 1,
+      "profile_id": "runtime-plugin",
+      "entry_digest": "1a6fd6f6a411eddb29e82d7b7abc7d5f658312f555dcf39e160a709194a8b867",
+      "status": "supported"
+    },
+    "0.14.2": {
+      "contract_id": "five-profile-v1",
+      "provenance_schema_version": 1,
+      "profile_id": "runtime-plugin",
+      "entry_digest": "f71ee26df3d4a89ecc727a1c30b40221cf062fce9f716a83453bced9503be8fd",
+      "status": "supported"
+    },
+    "0.14.3": {
+      "contract_id": "five-profile-v1",
+      "provenance_schema_version": 1,
+      "profile_id": "runtime-plugin",
+      "entry_digest": "c5dbe4044d11a976cd4ccd6de6cf1405ebfcb4a26ce5c6308c34f10142db50c9",
+      "status": "supported"
+    },
+    "0.14.4": {
+      "contract_id": "five-profile-v1",
+      "provenance_schema_version": 1,
+      "profile_id": "runtime-plugin",
+      "entry_digest": "a441d633fe11ea402c3366cfac0cec659d32eae5eb4ae6b1d25a1e366a029813",
+      "status": "supported"
+    }
+  }
+}

package/generated/plugins/agdf/runtime/create-agdf/generated/plugins/agdf/meta/agdf-plugin.definition.json

@@ -1,7 +1,7 @@
 {
   "schemaVersion": 1,
   "id": "agdf",
-  "version": "0.14.3",
+  "version": "0.14.4",
   "displayName": "AI Governance & Delivery Framework",
   "developerName": "Arndt Gold",
   "description": "Control layer for governed AI-assisted delivery.",

package/generated/plugins/agdf/runtime/create-agdf/lib/cli/command-registry.js

@@ -18,7 +18,7 @@ export const commandRegistry = Object.freeze([
   command("opencode-status", { preferred: [""], scaffold: [""] }),
   command("status", { preferred: [" [--surface <surface>] [--run <run_id>] [--json]"] }),
   command("runtime-checks", { preferred: [" <status|enable|manual> --surface <codex|claude|copilot|opencode> [--json]"] }),
-  command("disable", { preferred: [" --surface <surface> [--scope repository] [--dir <path>]"] }),
+  command("disable", { preferred: [" --surface <surface> [--scope repository] [--shared] [--dir <path>]"] }),
   command("uninstall", { preferred: [" --surface <surface> --scope global [--confirm]"] }),
   command("opencode-repo", { preferred: [""], scaffold: [""] }),
   command("init", { preferred: [""], scaffold: [""] }),
@@ -65,6 +65,12 @@ export function validateCommandOptions(options) {
     throw new Error("uninstall requires explicit --scope global");
   }
   if (options.confirm && options.target !== "uninstall") throw new Error("--confirm is supported only by uninstall");
+  if (options.shared && !(options.target === "disable" && options.surface === "copilot" && options.scope === "repository")) {
+    throw new Error("--shared is supported only by disable --surface copilot --scope repository");
+  }
+  if (options.target === "disable" && options.surface === "copilot" && options.scope !== "repository") {
+    throw new Error("Copilot disable requires explicit --scope repository");
+  }
   if (options.runtimeChecksDecision && !["codex", "claude", "copilot", "opencode"].includes(options.target)) {
     throw new Error("--runtime-checks is supported only by codex, claude, copilot and opencode installation commands");
   }
@@ -126,6 +132,7 @@ Options:
   --scope <repository|global>
                  Select the lifecycle mutation scope
   --confirm      Apply a previously previewed global uninstall plan
+  --shared       Apply Copilot repository disable through shared .github/copilot/settings.json
   --runtime-checks <enable|manual|cancel>
                  Make the installation-time automatic-check decision explicitly; no TTY defaults to manual
   --fixture <path>

package/generated/plugins/agdf/runtime/create-agdf/lib/cli/parse-args.js

@@ -45,6 +45,7 @@ export function parseArgs(argv, dependencies = {}) {
   let allActive = false;
   let scope;
   let confirm = false;
+  let shared = false;
   let runtimeChecksDecision;
   let runtimeChecksAction;
 
@@ -61,6 +62,7 @@ export function parseArgs(argv, dependencies = {}) {
     if (arg === "--generate-candidates") { generateCandidates = true; continue; }
     if (arg === "--all-active") { allActive = true; continue; }
     if (arg === "--confirm") { confirm = true; continue; }
+    if (arg === "--shared") { shared = true; continue; }
 
     if (arg === "--runtime-checks") {
       const next = requiredValue(args, i, arg);
@@ -162,6 +164,7 @@ export function parseArgs(argv, dependencies = {}) {
       allActive,
       scope,
       confirm,
+      shared,
       runtimeChecksDecision,
       runtimeChecksAction: runtimeChecksAction ?? "status",
       generatorModel,

package/generated/plugins/agdf/runtime/create-agdf/lib/control-evaluation/doctor.js

@@ -130,6 +130,19 @@ export function evaluateDoctor(targetDir, selection = {}) {
           "Add at least one evidence row or explicitly document that no evidence exists yet.",
         );
       }
+
+      for (const identityFinding of selectedRunState.identity_findings ?? []) {
+        addFinding(
+          findings,
+          "revise",
+          identityFinding.code,
+          identityFinding.code === "AGDF_RUN_ID_INVALID"
+            ? "The selected run state has a missing or invalid run_id; interaction cards cannot be rendered for it."
+            : "The selected run state has a missing or invalid revision_id; approval interactions cannot be rendered for it.",
+          runPath,
+          "Run run-migrate to repair the run identity, or migrate the legacy state to a canonical run record.",
+        );
+      }
     }
 
     const backlogPath = join(".agdf", "control", "MASTER_BACKLOG.md");

package/generated/plugins/agdf/runtime/create-agdf/lib/control-evaluation/gate-check.js

@@ -1,4 +1,4 @@
-import { attachApprovalOrientationSnapshot, buildArtefactRefs, buildQualityReadiness, gateTitle, localePack, renderApprovalOrientationSnapshot, renderOperationalStatusCard, resolveHumanRunTitle } from '../interaction-presentation.js';
+import { attachApprovalOrientationSnapshot, buildArtefactRefs, buildQualityReadiness, gateTitle, localePack, renderApprovalOrientationSnapshot, renderOperationalStatusCard, resolveHumanRunTitle, validateApprovalOrientationPreconditions, validateApprovalOrientationSnapshot, validateOperationalStatusCardPreconditions } from '../interaction-presentation.js';
 import { interactionLocales, resolveConfiguredChatLanguage } from '../cli/runtime-context.js';
 import { evaluateDoctor } from './doctor.js';
 import { analyzeDeliveryMap, deriveQualityOutlook } from './delivery-map.js';
@@ -300,8 +300,9 @@ export function evaluateGateCheck(targetDir, selection = {}) {
     humanPresentation,
     revisionId,
   });
+  const readyForApproval = isReadyUserGateApproval({ status, currentGate, missingApproval });
   const approvalOrientation = attachApprovalOrientationSnapshot(statusCard, {
-    ready: isReadyUserGateApproval({ status, currentGate, missingApproval }),
+    ready: readyForApproval,
     humanPresentation,
     revisionId,
     registry: interactionLocales,
@@ -312,6 +313,50 @@ export function evaluateGateCheck(targetDir, selection = {}) {
     enumerable: false,
   });
 
+  const approvalPresentation = renderApprovalOrientationSnapshot(approvalOrientation, {
+    registry: interactionLocales,
+    expectedIdentity: {
+      run_id: statusCard.run_id,
+      revision_id: revisionId,
+      current_gate: currentGate,
+      presentation_language: statusCard.presentation_language,
+    },
+  });
+  const presentationDiagnostics = {};
+  if (!statusPresentation) {
+    presentationDiagnostics.status_presentation_errors = [
+      ...validateOperationalStatusCardPreconditions(statusCard, {
+        registry: interactionLocales,
+        humanPresentation,
+      }).errors,
+    ];
+  }
+  if (readyForApproval && !approvalPresentation) {
+    presentationDiagnostics.approval_presentation_errors = approvalOrientation
+      ? [
+          ...validateApprovalOrientationSnapshot(approvalOrientation, {
+            registry: interactionLocales,
+            expectedIdentity: {
+              run_id: statusCard.run_id,
+              revision_id: revisionId,
+              current_gate: currentGate,
+              presentation_language: statusCard.presentation_language,
+            },
+          }).errors,
+        ]
+      : [
+          ...validateApprovalOrientationPreconditions({
+            statusCard,
+            humanPresentation,
+            registry: interactionLocales,
+            requestedLocale: presentationLocale,
+          }).errors,
+        ];
+    if (!presentationDiagnostics.approval_presentation_errors.length) {
+      presentationDiagnostics.approval_presentation_errors = ["snapshot_unavailable"];
+    }
+  }
+
   return {
     schema_version: "1",
     status,
@@ -332,15 +377,8 @@ export function evaluateGateCheck(targetDir, selection = {}) {
     candidate_runs: runState.candidate_runs ?? [],
     status_card: statusCard,
     status_presentation: statusPresentation,
-    approval_presentation: renderApprovalOrientationSnapshot(approvalOrientation, {
-      registry: interactionLocales,
-      expectedIdentity: {
-        run_id: statusCard.run_id,
-        revision_id: revisionId,
-        current_gate: currentGate,
-        presentation_language: statusCard.presentation_language,
-      },
-    }),
+    approval_presentation: approvalPresentation,
+    ...(Object.keys(presentationDiagnostics).length ? { presentation_diagnostics: presentationDiagnostics } : {}),
     delivery_map: {
       relationships: deliveryMap.relationships,
       mode_slice_decision: runState.mode_slice_decision,
@@ -368,6 +406,14 @@ export function printApprovalEnvelope(report, { io = console, reEvaluate } = {})
   if (report?.approval_presentation) {
     io.log(report.approval_presentation.blocks.run_status_card.markdown);
     io.log("");
+    if (report.status_presentation?.markdown) {
+      io.log(report.status_presentation.markdown);
+    } else {
+      const pack = localePack(interactionLocales, report.approval_presentation.presentation_language || "en");
+      const codes = report.presentation_diagnostics?.status_presentation_errors ?? [];
+      io.log(codes.length ? `${pack.interaction.statusPresentationFailure} (${codes.join(", ")})` : pack.interaction.statusPresentationFailure);
+    }
+    io.log("");
     io.log(report.approval_presentation.blocks.gate_transition_card.markdown);
     io.log("");
     io.log(report.approval_presentation.approval_interaction.exact_text_fallback);
@@ -386,7 +432,10 @@ export function printApprovalEnvelope(report, { io = console, reEvaluate } = {})
   const pack = localePack(interactionLocales, refreshed?.status_card?.presentation_language || report?.status_card?.presentation_language || "en");
   const refreshedApproval = canonicalApprovalForReport(refreshed);
   if (initialApproval && refreshedApproval) {
-    io.log(pack.interaction.presentationFailure);
+    const codes = refreshed?.presentation_diagnostics?.approval_presentation_errors
+      ?? report?.presentation_diagnostics?.approval_presentation_errors
+      ?? [];
+    io.log(codes.length ? `${pack.interaction.presentationFailure} (${codes.join(", ")})` : pack.interaction.presentationFailure);
     io.log(pack.interaction.exactTextRequest.replace("{approval}", refreshedApproval));
     return Object.freeze({ outcome: "exact_text_recovery", requested_decision: true, status: refreshed.status });
   }
@@ -404,7 +453,9 @@ export function printApprovalEnvelope(report, { io = console, reEvaluate } = {})
 function printGateCheckStatusCard(report, io) {
   const card = report.status_card;
   if (!report.status_presentation?.markdown) {
-    io.log(localePack(interactionLocales, card?.presentation_language || "en").interaction.statusPresentationFailure);
+    const failure = localePack(interactionLocales, card?.presentation_language || "en").interaction.statusPresentationFailure;
+    const codes = report.presentation_diagnostics?.status_presentation_errors ?? [];
+    io.log(codes.length ? `${failure} (${codes.join(", ")})` : failure);
     return false;
   }
   io.log(report.status_presentation.markdown);

package/generated/plugins/agdf/runtime/create-agdf/lib/control-evaluation/run-state.js

@@ -2,6 +2,7 @@ import { existsSync, readFileSync, realpathSync, statSync } from "node:fs";
 import { isAbsolute, join, relative, resolve } from "node:path";
 import process from "node:process";
 import { parseControlState, resolveRuns } from "../control-state/index.js";
+import { validateRunIdentity } from "../control-state/run-identity.js";
 import { buildRunCandidates } from "../interaction-presentation.js";
 import { cleanStatusCell, filled, isPlaceholderValue, readTargetFile } from "./shared.js";
 import { extractField, isSafeRepoRelativePath, readVerifiedChangeRecord } from "./verified-change.js";
@@ -87,6 +88,7 @@ export function readRunState(targetDir, selection = {}) {
       memory: {},
       candidate_runs: candidateRuns,
       resolution_error: resolutionError,
+      identity_findings: [],
     };
   }
 
@@ -101,6 +103,10 @@ export function readRunState(targetDir, selection = {}) {
     path: runPath,
     content,
     ...parsed,
+    identity_findings: validateRunIdentity({
+      runId: extractField(content, "run_id").replace(/^`|`$/g, ""),
+      revisionId: extractField(content, "revision_id").replace(/^`|`$/g, ""),
+    }),
   };
 }
 

package/generated/plugins/agdf/runtime/create-agdf/lib/control-state/run-identity.js

@@ -0,0 +1,9 @@
+export const RUN_ID_PATTERN = /^[a-z0-9][a-z0-9._-]{0,127}$/;
+export const REVISION_ID_PATTERN = /^[0-9a-f]{8}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{12}$/i;
+
+export function validateRunIdentity({ runId, revisionId } = {}) {
+  const findings = [];
+  if (!RUN_ID_PATTERN.test(String(runId ?? ""))) findings.push({ code: "AGDF_RUN_ID_INVALID" });
+  if (!REVISION_ID_PATTERN.test(String(revisionId ?? ""))) findings.push({ code: "AGDF_RUN_REVISION_ID_INVALID" });
+  return findings;
+}

package/generated/plugins/agdf/runtime/create-agdf/lib/control-state/run-state-parser.js

@@ -1,4 +1,6 @@
-export const RUN_ID_PATTERN = /^[a-z0-9][a-z0-9._-]{0,127}$/;
+import { REVISION_ID_PATTERN, RUN_ID_PATTERN } from "./run-identity.js";
+
+export { RUN_ID_PATTERN } from "./run-identity.js";
 export const LIFECYCLES = new Set([
   "active",
   "completed",
@@ -34,7 +36,7 @@ export function parseRunState(content, expected) {
     findings.push({ code: "AGDF_RUN_LIFECYCLE_INVALID" });
   if (!/^[1-9]\d*$/.test(values.get("revision") ?? ""))
     findings.push({ code: "AGDF_RUN_REVISION_INVALID" });
-  if (!/^[0-9a-f]{8}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{12}$/i.test(values.get("revision_id") ?? ""))
+  if (!REVISION_ID_PATTERN.test(values.get("revision_id") ?? ""))
     findings.push({ code: "AGDF_RUN_REVISION_ID_INVALID" });
   return {
     content,