memorix 1.2.1 → 1.2.3

package/docs/API_REFERENCE.md

@@ -35,6 +35,7 @@ Use the **CLI** when:
 The current CLI namespaces are:
 
 - `memorix session`
+- `memorix identity`
 - `memorix memory`
 - `memorix codegraph`
 - `memorix knowledge`
@@ -53,13 +54,16 @@ The current CLI namespaces are:
 - `memorix receipt`
 - `memorix sync`
 - `memorix ingest`
+- `memorix workbench`
 
 Typical examples:
 
 ```bash
-memorix session start --agent codex-main --agentType codex
+memorix --cwd /path/to/repo context --task "continue auth bug"
+memorix identity join --agent-type codex --name codex-main
+memorix session start --agent codex-main --agent-type codex --join-team --use
 memorix memory search --query "release blocker"
-memorix context --task "continue auth bug"
+memorix memory store --text "private triage note" --visibility personal
 memorix codegraph refresh
 memorix codegraph status --json
 memorix knowledge status
@@ -70,7 +74,7 @@ memorix task claim --taskId <id> --agentId <agent-id>
 memorix message inbox --agentId <agent-id>
 memorix lock status --file src/cli/index.ts
 memorix audit project
-memorix transfer export --format markdown
+memorix transfer export --format markdown --out ./memorix-export.md
 memorix skills show --name auth-pattern
 memorix sync workspace --action scan
 memorix ingest image --path ./diagram.png
@@ -78,7 +82,7 @@ memorix poll --agentId <agent-id>
 memorix receipt --json --probe "release blocker"
 ```
 
-The CLI is for direct terminal use, not a 1:1 mirror of MCP tool names. The only MCP-only area is the optional graph-compatibility tools (`create_entities`, `read_graph`, and related tools) for workflows that expect the official memory-server style graph API.
+The CLI is for direct terminal use, not a 1:1 mirror of MCP tool names. It does not require an MCP connection. `--cwd` selects a Git project from any shell; an unbound terminal has project-visible access only, including transfer exports. `memorix identity join|use|clear` makes personal/team access and coordination explicit, while `--as <agent-id>` is the one-command equivalent for scripts. The only MCP-only area is the optional graph-compatibility tools (`create_entities`, `read_graph`, and related tools) for workflows that expect the official memory-server style graph API.
 
 ### Memory Autopilot, Code State, and Context Packs
 
@@ -208,6 +212,13 @@ Important inputs:
 - optional `source`
 - optional `relatedCommits`
 - optional `relatedEntities`
+- optional `visibility`: `project` (default), `personal`, or `team`
+
+Visibility controls who can retrieve a record through agent-facing memory
+tools. Normal memories default to `project`. `personal` and `team` writes
+require a joined coordination identity; a personal record is readable only by
+its creator and explicitly named recipients. Supplying a `topicKey` never
+lets an agent overwrite a record outside its write scope.
 
 Example:
 
@@ -457,6 +468,10 @@ Important inputs:
 
 Export or import project memory.
 
+Exports contain only observations visible to the calling agent. An unbound
+agent receives project-visible observations only; personal and team records
+require its explicit active identity.
+
 Important inputs:
 
 - `action`
@@ -666,6 +681,11 @@ Important inputs:
 
 Use it when work should survive agent/session boundaries without relying on an IDE chat window staying alive.
 
+A targeted handoff is private to its sender and recipient. A handoff without
+`toAgentId` is visible only to active members of that project's team. Reading a
+targeted handoff does not give its recipient permission to rewrite or resolve
+it on behalf of the sender.
+
 ---
 
 ## 10. Ingestion Tools

package/docs/README.md

@@ -87,7 +87,7 @@ The public docs are organized by user intent:
 | 1.2 task-shaped evidence selection | [1.2 Workset Retrieval](1.2.0-WORKSET-RETRIEVAL.md) |
 | 1.2 non-blocking refresh and maintenance contract | [1.2 Dynamic Lifecycle](1.2.0-DYNAMIC-LIFECYCLE.md) |
 | 1.2 honest Lite and optional semantic CodeGraph provider contract | [1.2 Provider Quality](1.2.0-PROVIDER-QUALITY.md) |
-| Active 1.2 multi-dimensional memory work | [1.2.0 Development Charter](1.2.0-DEVELOPMENT-CHARTER.md) |
+| Active context-control work | [1.2.2 Memory Control Plane](1.2.2-MEMORY-CONTROL-PLANE.md) |
 | Historical cloud sync and multi-agent research | [CLOUD_SYNC_AND_MULTI_AGENT_RESEARCH.md](CLOUD_SYNC_AND_MULTI_AGENT_RESEARCH.md) |
 | Known issues and old roadmap notes | [KNOWN_ISSUES_AND_ROADMAP.md](KNOWN_ISSUES_AND_ROADMAP.md) |
 

package/docs/dev-log/progress.txt

@@ -4,15 +4,25 @@
 > older notes when they conflict.
 
 ## Current State
-- Phase: 1.2 release validation and publication
-- Branch: `codex/1.2.0-multidimensional-memory`
-- Last updated: 2026-07-18
-- Released baseline: `memorix@1.1.13`, tag `v1.1.13`, PR #129 merged as
-  `7e8077f`.
-- Goal: turn Memorix from a useful narrative-memory layer into a task-ready
-  working-context layer where code state, change evidence, verification,
-  source-backed project knowledge, workflows, and durable decisions are ranked
-  together.
+- Phase: 1.2.3 session-context visibility hotfix
+- Branch: `codex/1.2.3-session-visibility`
+- Last updated: 2026-07-25
+- Released baseline: `memorix@1.2.2`, tag `v1.2.2`, `origin/main` at
+  `2a7ebc7`.
+- Goal: close the agent-facing session-context visibility boundary before
+  further control-plane work.
+
+## 1.2.3 Session Context Visibility Hotfix
+- A real installed-package CLI smoke found that a second coordination identity
+  could receive the first identity's personal observation in automatic session
+  context. The root cause was a reader-less core session-context call.
+- `getSessionContext()` now accepts an `ObservationReader`; CLI, MCP, and TUI
+  session entry points pass their resolved caller identity. Alias-backed
+  project-visible records remain readable, while personal/team records fail
+  closed.
+- Graph-neighbor routing is omitted whenever a reader is present because graph
+  relations are not yet visibility-scoped. Core, CLI, MCP, and alias regression
+  tests cover the boundary.
 
 ## 1.1.13 Closeout
 - Codex setup now stamps plugin versions and doctor distinguishes bundle,
@@ -102,7 +112,7 @@
 - The compiled CLI was smoke-tested in an isolated Git project through init,
   compile, apply, and lint with a source-backed claim.
 
-## 1.2 Delivery Complete, Release Checks In Progress
+## 1.2 Delivery Complete, Released As 1.2.1
 - Phase 4 added canonical Markdown workflows, safe adapter preview/apply, import,
   selection, and verification receipts. Agent instruction files are render
   targets, not the workflow source of truth.
@@ -123,4 +133,84 @@
 - Release audit also incorporated #130 before publication: the optional SQLite
   runtime now uses the Node-26-compatible `better-sqlite3` 12.x line, with a
   dedicated Node 26 CI smoke that opens an in-memory database.
-- Remaining remote gates: GitHub CI, npm publish, tag, and GitHub release.
+- The release was completed as `memorix@1.2.1`. Its next development line is
+  [1.2.2 Memory Control Plane](../1.2.2-MEMORY-CONTROL-PLANE.md), which
+  consolidates these capabilities instead of adding another disconnected
+  memory feature.
+
+## 1.2.2 Control Plane
+- `TaskWorkset` already has bounded rendering and source-aware claims, but
+  SessionStart handoff and legacy context surfaces need one shared admission,
+  budget, and receipt contract.
+- First implementation slice complete: the shared `ContextReceipt` records
+  delivery target, selected source ids/kinds, trust/freshness, exact budget
+  omissions, queued refreshes, and Workset selection time without entering the
+  agent prompt. Project Context, Context Pack, and full SessionStart injection
+  now carry the same delivery-target semantics; `memorix explain` exposes the
+  receipt in text and JSON.
+- Budget selection now keeps a knowledge page dependent on a Claim out of the
+  prompt when its qualifying Claim could not fit. The receipt records both as
+  withheld instead of presenting an orphaned page path as usable knowledge.
+- Phase 3 admission slice complete: automatic observations now persist an
+  `ephemeral`, `candidate`, or `qualified` state and a reason. Hooks and the
+  internal orchestrator capture concrete evidence as candidates, keep routine
+  success/telemetry ephemeral, and drop low-signal activity without writing it.
+  Automatic capture stays quiet in the host agent.
+- A deduplicated `observation-qualify` maintenance job runs after CodeGraph
+  refresh. It requires a scan no older than the capture and at least one
+  current code reference before upgrading a candidate. Qualification does not
+  create a Claim; explicit saves and Git remain the Claim boundary.
+- Auto Project Context, automatic session L1 delivery, Wiki/Knowledge Graph
+  compilation, and graph-context packets exclude candidates and ephemeral
+  traces. Ordinary search prefers qualified material, while explicit lookup
+  can still inspect pending evidence.
+- Pending automatic evidence is also excluded from automatic consolidation so
+  a later qualification can still inspect its original source grain. Retention
+  explanations no longer describe an unqualified `core` candidate as immune.
+- Actor and visibility enforcement is now one policy shared by MCP retrieval,
+  auto context, CLI/TUI, dashboards, knowledge/graph projections, and scoped
+  maintenance. Legacy observations remain project-visible; new records may be
+  `project`, `team`, or `personal` with an explicit creator and recipients.
+  Missing identity never grants private or team visibility.
+- Automatic hooks and orchestrator traces begin personal. A verified hook
+  candidate may become project evidence only when qualification proves a
+  current code reference. Targeted handoffs are readable by sender and
+  recipient but writable only by sender; broadcasts require an active project
+  team member.
+- Automatic orchestration lesson lookup is explicitly project-scoped, so its
+  prompt injection cannot read personal or team-scoped records.
+- Topic-key writes now check mutation scope before storage, and the
+  project-attribution guard receives only memories visible to its current
+  reader. This closes both guessed-key overwrite and cross-project existence
+  hints without hiding historical project knowledge after upgrade.
+- The embedded SDK remains project-visible when unbound. CLI and the local TUI
+  now share the same explicit actor reader: an unbound terminal sees project
+  memory only, while `memorix identity join|use` (or one-shot `--as`) grants
+  only that active project member's personal/team visibility and coordination
+  ergonomics. `identity clear` returns to project scope; private evidence
+  cannot be promoted into shared skills or generated project skills.
+- CLI control-plane ergonomics now include `--cwd`, kebab-case flag aliases,
+  action-oriented namespace help, explicit `memorix workbench`, and
+  file/stdin transfer import/export. Root memory aliases use the same
+  canonical command paths as `memorix memory`; CLI and MCP exports are filtered
+  by the same active reader so an unbound backup cannot contain private/team
+  observations.
+- Direct CLI and foreground Code Memory refreshes now enqueue qualification as
+  well as claim requalification, and they register the local project target
+  needed by the isolated control-plane worker. The host hook path persists its
+  candidate before scheduling a refresh, avoiding a fast-worker race where a
+  scan could predate the capture.
+- Compiled-artifact smoke: an isolated Codex `PostToolUse` hook returned only
+  `continue: true`, stored a `candidate`, and the isolated background control
+  plane upgraded it to `qualified` after a real Code Memory scan with one
+  current reference. Embeddings were deliberately disabled for this
+  deterministic local smoke. The full test suite, TypeScript check, and
+  production build also passed after the admission policy was reconciled with
+  OpenCode's concise technical response events.
+- #135 remains a contributor-owned session-handoff candidate for the bounded
+  `HandoffPacket` phase. It is not copied or silently replaced.
+- #133/#134 remain open pending an official provider-capability layer and a
+  safe agent-facing media path. The existing OpenRouter text embedding path is
+  the supported baseline.
+- See [1.2.2 Memory Control Plane](../1.2.2-MEMORY-CONTROL-PLANE.md) for
+  phases, acceptance gates, non-goals, and release criteria.

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "memorix",
-  "version": "1.2.1",
+  "version": "1.2.3",
   "description": "Local-first shared memory layer for AI coding agents across MCP clients, Git history, reasoning context, and project sessions.",
   "type": "module",
   "sideEffects": false,

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

@@ -1,6 +1,6 @@
 {
   "name": "memorix",
-  "version": "1.2.1",
+  "version": "1.2.3",
   "description": "Shared workspace memory for Codex and other AI coding agents.",
   "author": {
     "name": "AVIDS2",

package/src/cli/command-guide.ts

@@ -0,0 +1,192 @@
+export interface CliCommandGuide {
+  summary: string;
+  usage: string[];
+  notes?: string[];
+}
+
+const GUIDES: Record<string, CliCommandGuide> = {
+  memory: {
+    summary: 'Store, search, inspect, resolve, consolidate, and promote project memory.',
+    usage: [
+      'memorix memory search --query "timeout regression" [--limit 10]',
+      'memorix memory store --text "..." [--title "..."] [--visibility project|personal|team]',
+      'memorix memory detail --id 42',
+      'memorix memory recent [--limit 10]',
+      'memorix memory resolve --ids 42,43 [--status resolved|archived]',
+      'memorix memory consolidate --action preview|execute',
+      'memorix memory promote --ids 42,43 [--trigger "..."] [--instruction "..."]',
+    ],
+    notes: ['Personal and team visibility require `memorix identity join` or `memorix identity use` first.'],
+  },
+  identity: {
+    summary: 'Select the active local actor for private memory and coordination commands.',
+    usage: [
+      'memorix identity status',
+      'memorix identity join --agent-type codex [--name codex-main --instance-id local]',
+      'memorix identity use --agent-id <id>',
+      'memorix identity clear',
+    ],
+    notes: ['Without an identity, CLI reads and writes only project-visible memory.'],
+  },
+  session: {
+    summary: 'Start, end, and inspect coding sessions. Coordination remains opt-in.',
+    usage: [
+      'memorix session start --agent codex',
+      'memorix session start --agent codex --agent-type codex --join-team --use',
+      'memorix session end --session-id <id> [--summary "..."]',
+      'memorix session context [--limit 3]',
+    ],
+  },
+  context: {
+    summary: 'Build the bounded task Workset used to resume or begin real work.',
+    usage: ['memorix context --task "continue the release fix" [--refresh auto|always|never]'],
+  },
+  explain: {
+    summary: 'Show why a task context contains its current facts and memory evidence.',
+    usage: ['memorix explain [--refresh auto|always|never]'],
+  },
+  codegraph: {
+    summary: 'Refresh local code-state snapshots and assemble task-specific code context.',
+    usage: [
+      'memorix codegraph status',
+      'memorix codegraph refresh',
+      'memorix codegraph context-pack --task "trace the auth flow" [--limit 20]',
+    ],
+  },
+  knowledge: {
+    summary: 'Manage the reviewed, source-backed Knowledge Workspace and project workflows.',
+    usage: [
+      'memorix knowledge init [--mode local|versioned]',
+      'memorix knowledge status',
+      'memorix knowledge compile',
+      'memorix knowledge lint',
+      'memorix knowledge workflow list',
+    ],
+  },
+  reasoning: {
+    summary: 'Capture and retrieve engineering decision rationale.',
+    usage: [
+      'memorix reasoning store --entity auth --decision "Use SQLite" --rationale "..."',
+      'memorix reasoning search --query "why sqlite" [--scope project|global]',
+    ],
+  },
+  retention: {
+    summary: 'Inspect retention state and archive records that are eligible under the current scope.',
+    usage: [
+      'memorix retention status',
+      'memorix retention stale',
+      'memorix retention archive',
+    ],
+  },
+  transfer: {
+    summary: 'Create or restore explicit project memory snapshots for backup and automation.',
+    usage: [
+      'memorix transfer export --format json --out ./.memorix-export.json',
+      'memorix transfer import --file ./.memorix-export.json',
+      'memorix transfer import --stdin',
+    ],
+  },
+  team: {
+    summary: 'Manage explicit multi-agent coordination state for the current project.',
+    usage: [
+      'memorix team join --agent-type codex [--name codex-main]',
+      'memorix team status [--all]',
+      'memorix team leave [--agent-id <id>]',
+    ],
+  },
+  task: {
+    summary: 'Create and progress coordination tasks. An active CLI identity fills agentId automatically.',
+    usage: [
+      'memorix task create --description "Review release evidence"',
+      'memorix task claim --task-id <id>',
+      'memorix task complete --task-id <id> --result "Verified"',
+      'memorix task list [--available]',
+    ],
+  },
+  message: {
+    summary: 'Send and read coordination messages. An active CLI identity fills the sender automatically.',
+    usage: [
+      'memorix message send --to <agent-id> --type info --content "..."',
+      'memorix message broadcast --type announcement --content "..."',
+      'memorix message inbox [--mark-read]',
+    ],
+  },
+  lock: {
+    summary: 'Use advisory file locks to reduce overlapping edits between agents.',
+    usage: [
+      'memorix lock lock --file src/server.ts',
+      'memorix lock unlock --file src/server.ts',
+      'memorix lock status [--file src/server.ts]',
+    ],
+  },
+  handoff: {
+    summary: 'Create durable, targeted or team-visible handoff artifacts.',
+    usage: ['memorix handoff send --summary "..." --context "..." [--to-agent-id <id>]'],
+  },
+  poll: {
+    summary: 'Return a compact coordination snapshot for the active CLI identity.',
+    usage: ['memorix poll [--mark-inbox-read]'],
+  },
+  audit: {
+    summary: 'Inspect quality, attribution, and recorded audit evidence.',
+    usage: ['memorix audit memory', 'memorix audit project [--threshold 2]', 'memorix audit list'],
+  },
+  skills: {
+    summary: 'Inspect and generate project skills from project-visible memory evidence.',
+    usage: ['memorix skills list', 'memorix skills generate [--write --target codex]', 'memorix skills show --name <skill>'],
+  },
+  sync: {
+    summary: 'Synchronize agent rules and workspace artifacts through explicit routes.',
+    usage: ['memorix sync rules --action status', 'memorix sync workspace --action scan'],
+  },
+  ingest: {
+    summary: 'Ingest Git and image evidence into project memory.',
+    usage: ['memorix ingest commit [--ref HEAD]', 'memorix ingest log [--count 10]', 'memorix ingest image --path ./diagram.png'],
+  },
+  workbench: {
+    summary: 'Open the interactive terminal memory control plane.',
+    usage: ['memorix workbench'],
+  },
+};
+
+export function commandGuideNames(): string[] {
+  return Object.keys(GUIDES).sort();
+}
+
+export function renderCliGuide(command?: string): string {
+  const normalized = command?.trim().toLowerCase();
+  if (normalized && GUIDES[normalized]) {
+    const guide = GUIDES[normalized];
+    return [
+      `Memorix ${normalized}`,
+      '',
+      guide.summary,
+      '',
+      'Usage:',
+      ...guide.usage.map((line) => `  ${line}`),
+      ...(guide.notes?.length ? ['', ...guide.notes.map((line) => `Note: ${line}`)] : []),
+      '',
+      'Project/actor options: --cwd <git-project>  --as <active-agent-id>. Operator commands accept --json.',
+    ].join('\n');
+  }
+
+  return [
+    'Memorix CLI',
+    '',
+    'Use `memorix <command> --help` for an action-oriented guide.',
+    '',
+    `Command groups: ${commandGuideNames().join(', ')}`,
+    '',
+    'Global options: --cwd <git-project>  --as <active-agent-id>',
+    'Compatibility: camelCase flags remain supported; kebab-case flags are accepted too.',
+  ].join('\n');
+}
+
+/** Render manual action help before Citty consumes `--help` as flag metadata. */
+export function printCliGuideForHelp(argv: string[] = process.argv.slice(2)): boolean {
+  const [command, ...rest] = argv;
+  if (!command || !GUIDES[command.toLowerCase()]) return false;
+  if (!rest.includes('--help') && !rest.includes('-h')) return false;
+  console.log(renderCliGuide(command));
+  return true;
+}

package/src/cli/commands/audit.ts

@@ -3,6 +3,7 @@ import { getAllAuditEntries, getProjectId } from '../../audit/index.js';
 import { auditProjectObservations } from '../../memory/attribution-guard.js';
 import { getAllObservations } from '../../memory/observations.js';
 import { auditMemoryQuality } from '../../memory/quality-audit.js';
+import { filterReadableObservations } from '../../memory/visibility.js';
 import { emitError, emitResult, getCliProjectContext } from './operator-shared.js';
 
 export default defineCommand({
@@ -22,8 +23,8 @@ export default defineCommand({
     try {
       switch (action) {
         case 'memory': {
-          const { project } = await getCliProjectContext();
-          const report = auditMemoryQuality(getAllObservations(), {
+          const { project, reader } = await getCliProjectContext();
+          const report = auditMemoryQuality(filterReadableObservations(getAllObservations(), reader), {
             projectId: project.id,
           });
           emitResult(
@@ -68,9 +69,13 @@ export default defineCommand({
         }
 
         case 'project': {
-          const { project } = await getCliProjectContext();
+          const { project, reader } = await getCliProjectContext();
           const threshold = Number.parseInt(String(args.threshold ?? '2'), 10) || 2;
-          const entries = await auditProjectObservations(project.id, getAllObservations(), threshold);
+          const entries = await auditProjectObservations(
+            project.id,
+            filterReadableObservations(getAllObservations(), reader),
+            threshold,
+          );
           emitResult(
             { project, entries, threshold },
             entries.length === 0

package/src/cli/commands/cleanup.ts

@@ -19,6 +19,7 @@ import { detectProject } from '../../project/detector.js';
 import { getProjectDataDir } from '../../store/persistence.js';
 import type { ObservationStore } from '../../store/obs-store.js';
 import { getObservationStore, initObservationStore } from '../../store/obs-store.js';
+import { filterReadableObservations } from '../../memory/visibility.js';
 
 /** Patterns that indicate auto-generated, low-value observations */
 const LOW_QUALITY_PATTERNS = [
@@ -148,7 +149,10 @@ export default defineCommand({
         const dataDir = await getProjectDataDir(projectId);
         await initObservationStore(dataDir);
         const store = getObservationStore();
-        const projectObs = await store.loadByProject(projectId, { status: 'active' }) as Array<{
+        const projectObs = filterReadableObservations(
+            await store.loadByProject(projectId, { status: 'active' }),
+            { projectId },
+        ) as Array<{
             id?: number;
             type?: string;
             title?: string;

package/src/cli/commands/codegraph.ts

@@ -9,6 +9,7 @@ import { getExternalCodeGraphContext, inspectExternalCodeGraph } from '../../cod
 import type { CodeGraphProviderQuality } from '../../codegraph/types.js';
 import { getResolvedConfig } from '../../config/resolved-config.js';
 import { getAllObservations } from '../../memory/observations.js';
+import { filterReadableObservations } from '../../memory/visibility.js';
 import { emitError, emitResult, getCliProjectContext, parsePositiveInt } from './operator-shared.js';
 
 function formatSnapshotStatus(status: ReturnType<CodeGraphStore['status']>): string[] {
@@ -92,7 +93,7 @@ export default defineCommand({
     const asJson = !!args.json;
 
     try {
-      const { project, dataDir } = await getCliProjectContext();
+      const { project, dataDir, reader } = await getCliProjectContext();
       const store = new CodeGraphStore();
       await store.init(dataDir);
       const explicitAction = Boolean(positional[0] || (args.action as string | undefined));
@@ -125,13 +126,21 @@ export default defineCommand({
           const activeObservations = getAllObservations()
             .filter(obs => obs.projectId === project.id && (obs.status ?? 'active') === 'active');
           const backfill = await backfillMissingObservationCodeRefs(store, activeObservations);
-          const { enqueueClaimRequalification } = await import('../../runtime/lifecycle.js');
+          const {
+            enqueueClaimRequalification,
+            enqueueObservationQualification,
+          } = await import('../../runtime/lifecycle.js');
           enqueueClaimRequalification({
             dataDir,
             projectId: project.id,
             source: 'manual-codegraph-refresh',
             snapshotId: refresh.snapshot.id,
           });
+          enqueueObservationQualification({
+            dataDir,
+            projectId: project.id,
+            source: 'manual-codegraph-refresh',
+          });
           const status = store.status(project.id);
           const providerQuality = await inspectExternalCodeGraph({
             projectRoot: project.rootPath,
@@ -160,9 +169,10 @@ export default defineCommand({
             return;
           }
           const limit = parsePositiveInt(args.limit as string | undefined, 20);
-          const observations = getAllObservations()
-            .filter(obs => obs.projectId === project.id && (obs.status ?? 'active') === 'active')
-            .reverse();
+          const observations = filterReadableObservations(
+            getAllObservations().filter(obs => obs.projectId === project.id && (obs.status ?? 'active') === 'active'),
+            reader,
+          ).reverse();
           const basePack = assembleContextPackForTask({
             store,
             projectId: project.id,

package/src/cli/commands/context.ts

@@ -6,6 +6,7 @@ import {
   type AutoContextRefreshMode,
 } from '../../codegraph/auto-context.js';
 import { getAllObservations } from '../../memory/observations.js';
+import { filterReadableObservations } from '../../memory/visibility.js';
 import { emitError, emitResult, getCliProjectContext } from './operator-shared.js';
 
 function coerceRefreshMode(input?: string): AutoContextRefreshMode {
@@ -28,11 +29,11 @@ export default defineCommand({
     const asJson = !!args.json;
 
     try {
-      const { project, dataDir } = await getCliProjectContext();
+      const { project, dataDir, reader } = await getCliProjectContext();
       const context = await buildAutoProjectContext({
         project,
         dataDir,
-        observations: getAllObservations(),
+        observations: filterReadableObservations(getAllObservations(), reader),
         task: args.task as string | undefined,
         refresh: coerceRefreshMode(args.refresh as string | undefined),
       });

package/src/cli/commands/doctor.ts

@@ -257,8 +257,10 @@ export default defineCommand({
         await initObservationStore(dataDir);
         const store = getObservationStore();
         backendName = store.getBackendName();
-        const obs = await store.loadAll();
-        projectObservations = obs.filter((o: any) => o.projectId === projectId);
+        const { filterReadableObservations } = await import('../../memory/visibility.js');
+        projectObservations = projectId
+          ? filterReadableObservations(await store.loadAll(), { projectId })
+          : [];
         obsCount = projectObservations.length;
         activeCount = projectObservations.filter((o: any) => (o.status ?? 'active') === 'active').length;
         ranCount = projectObservations.filter((o: any) => /^Ran:\s/i.test(o.title ?? '')).length;

package/src/cli/commands/explain.ts

@@ -1,7 +1,9 @@
 import { defineCommand } from 'citty';
 import { buildAutoProjectContext, type AutoContextRefreshMode } from '../../codegraph/auto-context.js';
 import { formatProjectContextExplain } from '../../codegraph/project-context.js';
+import { formatContextReceipt } from '../../knowledge/context-assembly.js';
 import { getAllObservations } from '../../memory/observations.js';
+import { filterReadableObservations } from '../../memory/visibility.js';
 import { emitError, emitResult, getCliProjectContext } from './operator-shared.js';
 
 function coerceRefreshMode(input?: string): AutoContextRefreshMode {
@@ -23,15 +25,19 @@ export default defineCommand({
     const asJson = !!args.json;
 
     try {
-      const { project, dataDir } = await getCliProjectContext();
+      const { project, dataDir, reader } = await getCliProjectContext();
       const context = await buildAutoProjectContext({
         project,
         dataDir,
-        observations: getAllObservations(),
+        observations: filterReadableObservations(getAllObservations(), reader),
         refresh: coerceRefreshMode(args.refresh as string | undefined),
       });
 
-      emitResult({ project, explain: context.explain, refresh: context.refresh }, formatProjectContextExplain(context.explain), asJson);
+      emitResult(
+        { project, explain: context.explain, receipt: context.workset.receipt, refresh: context.refresh },
+        formatProjectContextExplain(context.explain) + '\n\n' + formatContextReceipt(context.workset.receipt),
+        asJson,
+      );
     } catch (error) {
       emitError(error instanceof Error ? error.message : String(error), asJson);
     }