@a-company/paradigm 5.38.0 → 6.0.2

package/dist/university-content/notes/N-para-501-lore-system.md

@@ -0,0 +1,155 @@
+---
+id: N-para-501-lore-system
+title: The Lore System
+type: note
+author: paradigm
+created: '2026-04-22'
+updated: '2026-04-18'
+tags:
+  - course
+  - para-501
+  - lore-entries-record
+  - seven-entry-types
+  - date-partitioned-yaml-storage
+symbols: []
+difficulty: beginner
+estimatedMinutes: 5
+prerequisites: []
+category: paradigm-core
+origin: imported
+source: courses/para-501.json
+---
+
+## Why Projects Forget
+
+Every software project accumulates institutional knowledge — why a migration was attempted then rolled back, which approach was chosen for caching and why, what the team learned when the billing system went down at 2 AM. Without a system for capturing this knowledge, it lives only in the heads of the people who were there. When they leave, context-switch, or simply forget, the project loses its memory.
+
+Paradigm's Lore system is a structured project timeline. It records sessions, milestones, incidents, retros, reviews, and insights as date-partitioned YAML entries that both humans and AI agents can search, filter, and learn from.
+
+> **v6.0 change:** the `decision` lore type was removed. Architectural decisions now have their own dedicated store — see "Decisions Have Their Own Store" below.
+
+## Anatomy of a Lore Entry
+
+Every lore entry follows a consistent structure:
+
+```yaml
+id: L-2026-02-21-ascend-143025-001
+type: agent-session
+timestamp: "2026-02-21T14:30:25Z"
+duration_minutes: 45
+author: ascend
+agent:
+  provider: anthropic
+  model: claude-opus-4-6
+title: "Add JWT authentication to user routes"
+summary: "Implemented RS256 JWT auth middleware, added ^authenticated and ^project-admin gates to portal.yaml, created refresh token rotation."
+symbols_touched: ["#auth-middleware", "^authenticated", "^project-admin"]
+symbols_created: ["#refresh-token-handler"]
+files_modified: ["src/middleware/auth.ts", "portal.yaml"]
+files_created: ["src/handlers/refresh-token.ts"]
+lines_added: 247
+lines_removed: 12
+commit: "a1b2c3d"
+decisions:
+  - id: jwt-signing
+    decision: "Use RS256 over HS256"
+    rationale: "Allows public key verification without sharing the signing secret"
+learnings:
+  - "Express v5 requires explicit async error wrapping for middleware"
+verification:
+  status: pass
+  details: { "unit-tests": pass, "integration": pass }
+tags: [security, auth]
+```
+
+The `id` field is auto-generated as `L-{date}-{author}-{hhmmss}-{seq}` — date partitions the storage, author and timestamp disambiguate within a day, and the sequence handles burst writes. Note that `author` is the human user (a string), and AI assistance is recorded separately in the optional `agent` object. The `decisions` field on an agent-session entry remains valid — it captures decisions made *during* a work session. Standalone architectural decisions go through `paradigm_decision_record` (see below).
+
+## Entry Types
+
+Lore recognizes seven entry types, each capturing a different kind of project event:
+
+| Type | When to Use |
+|---|---|
+| `agent-session` | An AI agent completed a work session (most common) |
+| `human-note` | A human records context, rationale, or tribal knowledge |
+| `review` | A code review, PR review, or post-mortem |
+| `incident` | A production incident or significant failure |
+| `milestone` | A release, launch, migration completion, or major achievement |
+| `retro` | A retrospective on a sprint, project, or incident response |
+| `insight` | A learned pattern or observation worth preserving |
+
+The type drives how the entry appears in timeline views and which filters surface it.
+
+## Decisions Have Their Own Store
+
+In v6.0 the `decision` lore type was removed. Decisions are first-class enough to deserve their own store, separate from the time-partitioned narrative of lore. Decisions live in `.paradigm/decisions/` as `TD-*` entries, recorded via `paradigm_decision_record`. When you record a decision, Paradigm auto-writes a companion lore `insight` entry pointing at it (with `references.decision_id`) — so the timeline still surfaces the moment the decision was made, while the canonical decision record stays addressable by topic rather than by date.
+
+If you have a v1/v2 project with old `type: decision` lore entries, the storage layer migrates them to type `insight` on read and tags them `v6-migrated:from-decision` for forensic recovery. New entries with `type: 'decision'` are rejected at the storage layer with an error pointing you at the decision-record path. The rejection envelope (`code: 'lore_type_decision_removed'`, `successor_tool: 'paradigm_decision_record'`) is structured so calling agents can auto-retry without human intervention.
+
+Search hierarchy:
+- Use `paradigm_lore_search` for narrative ("what happened on 2026-02-21?").
+- Use `paradigm_decision_search` for canonical choices ("what did we decide about caching?").
+
+## Storage: Date-Partitioned YAML
+
+Lore entries live in `.paradigm/lore/entries/` organized by date. Filenames mirror the entry id (`L-{date}-{author}-{hhmmss}-{seq}.yaml`), so you can identify the author and burst order from the filename alone:
+
+```
+.paradigm/lore/
+  timeline.yaml          # Index metadata
+  entries/
+    2026-02-19/
+      L-2026-02-19-ascend-091203-001.yaml
+      L-2026-02-19-matt-143812-001.yaml
+    2026-02-20/
+      L-2026-02-20-ascend-101545-001.yaml
+    2026-02-21/
+      L-2026-02-21-ascend-143025-001.yaml
+```
+
+The `timeline.yaml` index tracks total entry count, last updated timestamp, and known authors. Date partitioning keeps directories small and makes time-range queries efficient — to find entries from last week, you only read 7 directories.
+
+## CLI Tools
+
+The CLI provides full lore management:
+
+- `paradigm lore list` — List entries with filters (author, type, symbol, date range, tags)
+- `paradigm lore show <id>` — Full detail view of a single entry
+- `paradigm lore record` — Record a new entry with expanded fields (files-modified, files-created, commit, learnings, duration)
+- `paradigm lore edit <id>` — Edit entry fields (title, summary, type, symbols, tags, learnings)
+- `paradigm lore delete <id>` — Delete an entry (with --yes to skip confirmation)
+- `paradigm lore timeline` — Timeline view grouped by date with hot symbols
+- `paradigm lore review <id>` — Add review scores to an entry
+- `paradigm lore` — Launch the web timeline UI
+
+## MCP Tools
+
+The Lore system exposes the following MCP tools:
+
+**`paradigm_lore_record`** — Create a new entry. Requires `type`, `title`, `summary`, and `symbols_touched`. Optional fields include files, decisions, learnings, and verification status. The entry is written to the correct date directory with an auto-incremented ID. When `validateSymbols: true` is passed, the tool checks each symbol in `symbols_touched` against registered symbols in `.purpose` files, `flows.yaml`, and `portal.yaml`. Unregistered symbols produce advisory warnings (the entry is always recorded regardless). Calling with `type: 'decision'` returns a structured rejection envelope pointing at `paradigm_decision_record`.
+
+**`paradigm_lore_search`** — Query entries with filters: by symbol, author, type, date range, tags, review status, and minimum completeness score. Returns matching entries sorted by recency.
+
+**`paradigm_lore_timeline`** — Get a high-level view: recent entries, active authors, hot symbols (most-referenced in recent entries), and timeline metadata. Use this for orientation — it tells you what has been happening in the project.
+
+**`paradigm_lore_get`** — Fetch a single entry by ID. Returns the full entry with all fields, including decisions, learnings, and review data.
+
+**`paradigm_lore_update`** — Update an existing entry. Pass the entry ID and the fields to change (title, summary, type, symbols, tags, learnings). Only specified fields are modified.
+
+**`paradigm_lore_delete`** — Delete an entry by ID. Requires `confirm: true` to prevent accidental deletion.
+
+**`paradigm_lore_assess`** — Score an entry's quality and completeness (1-5 each), with optional notes. The assessment is stored alongside the entry and feeds the calibration and review filters.
+
+**`paradigm_lore_calibration`** — Surface calibration drift: entries where the recorded confidence diverges from later-observed reality. Use this to find sessions where an agent over- or under-estimated its certainty.
+
+## Lore Reviews
+
+Entries can be reviewed by humans after the fact. A review adds a `completeness` score (1-5), a `quality` score (1-5), and optional notes. This creates a feedback loop: agents learn which sessions produced high-quality entries and can adjust their recording behavior. You can filter entries by `hasReview` and `minCompleteness` to surface only verified project history.
+
+## When to Record
+
+The general rule: **record lore when a session modifies 3 or more source files**. This threshold captures significant work sessions while ignoring trivial edits. The stop hook enforces this — if you modified 3+ files without recording a lore entry, it will block your session from completing.
+
+Beyond the threshold, always record lore for: production incidents, milestone completions, retros, and any session where you learned something the next developer should know. For standalone architectural decisions, use `paradigm_decision_record` instead — the decision store handles them, and a companion lore `insight` is auto-written for timeline coverage.
+
+> **Going deeper:** PARA 601 covers knowledge streams (work-log, journal, decisions) and how `paradigm_decision_record` writes to the decisions stream while emitting a companion lore `insight` for timeline coverage. PARA 301 covers the new dedicated decision store in detail (`N-para-301-decisions`).

package/dist/university-content/notes/N-para-501-platform-agent-ui.md

@@ -0,0 +1,108 @@
+---
+id: N-para-501-platform-agent-ui
+title: Platform & Agent-Driven UI
+type: note
+author: paradigm
+created: '2026-04-22'
+updated: '2026-04-22'
+tags:
+  - course
+  - para-501
+  - paradigm-serve-unifies
+  - agent-driven-ui-5
+  - pipeline-mcp-
+symbols: []
+difficulty: beginner
+estimatedMinutes: 3
+prerequisites: []
+category: paradigm-core
+origin: imported
+source: courses/para-501.json
+---
+
+## The Unified Platform
+
+`paradigm serve` launches the Paradigm Platform — a unified development management interface on port 3850 that absorbs every Paradigm tool (Lore, Graph, Sentinel, University, Symphony) into one browser tab.
+
+The Platform is built on Express + WebSocket on the server, React 18 + Zustand on the client. Sections are lazy-loaded. A shared design system provides consistent theming and symbol colors.
+
+### Architecture
+
+```
+localhost:3850 (Express + WebSocket)
+├── /api/lore/*        ← LoreRouter
+├── /api/symbols/*     ← SymbolsRouter
+├── /api/graphs/*      ← GraphsRouter
+├── /api/platform/*    ← PlatformRouter (health, sections, agent-command)
+├── /ws                ← WebSocket (agent commands + user activity)
+└── /                  ← Platform UI SPA
+```
+
+## Agent-Driven UI
+
+The breakthrough: **the AI agent can drive the browser in real-time.** Five MCP tools let the agent navigate, highlight, annotate, observe, and clear — turning the Platform from a passive viewer into a shared workspace.
+
+### The Pipeline: MCP → HTTP → WebSocket → Browser
+
+```
+Agent (Claude Code)       Platform Server          Browser
+      │                          │                    │
+      │ paradigm_platform_*      │                    │
+      │ POST /api/platform/cmd   │                    │
+      │ ─────────────────────────►│                    │
+      │   ◄── { ok: true } ──────│                    │
+      │                          │  ws: agent:*       │
+      │                          │──────────────────►│
+      │                          │                    │ UI updates
+```
+
+Why HTTP not file-based: the <500ms latency requirement rules out file-watching. Why not direct WebSocket from MCP: MCP tools are stdio-based with no event loop for persistent WS connections.
+
+### The Five Tools
+
+| Tool | Purpose |
+|------|---------|
+| `paradigm_platform_navigate` | Switch sections, select symbols, open lore entries |
+| `paradigm_platform_highlight` | Pulsing glow on symbols with color + label, auto-expires |
+| `paradigm_platform_annotate` | Toasts (notifications), callouts (on graph nodes), badges |
+| `paradigm_platform_observe` | Read user's current section, selected symbol, theme, mute state |
+| `paradigm_platform_clear` | Remove all agent highlights and annotations |
+
+### Conflict Resolution: User Always Wins
+
+The agent must never hijack the user's attention:
+
+- **User idle (>5s):** Agent navigation executes immediately
+- **User active (<5s):** A prompt appears: "Agent wants to show you #X — [Go there] [Dismiss]"
+- **User muted:** All agent effects are silently discarded; `observe` returns `{ muted: true }`
+
+### Agent Presence
+
+The `#AgentPresenceManager` tracks connected agents by their Symphony identity (`{project}/{role}`). Each agent gets a deterministic color from its ID hash. Presence dots appear in the Platform header with a mute toggle.
+
+Stale agents are auto-pruned after 2 minutes of inactivity.
+
+### User State Tracking
+
+The `#UserStateTracker` accumulates user activity — what section they're viewing, what symbol is selected, theme preference. This state is served to `paradigm_platform_observe` so the agent can reason about what the user is looking at.
+
+Browser clients report activity via WebSocket messages: `user:navigate`, `user:select`, `user:theme`, `user:mute`.
+
+### Visual Treatment
+
+| Element | Human | Agent |
+|---------|-------|-------|
+| Selection ring | Solid 2px blue | Dashed 2px agent-color |
+| Highlight | N/A | Pulsing glow animation |
+| Toast | N/A | Left border + robot icon |
+| Navigation | Instant | 300ms ease + toast notification |
+
+### Browser Architecture
+
+The agent UI layer sits alongside existing stores:
+
+- `agentStore.ts` — Zustand store managing presence, highlights, annotations, toasts, mute, pending navigation
+- `useAgentEffects` — Hook connecting WebSocket `agent:*` messages to store actions, with auto-reconnect
+- `useActivityReporter` — Hook reporting section/theme changes back to server
+- `AgentToast` — Severity-colored toast component
+- `AgentCallout` — Floating callout overlay + navigation conflict prompt

package/dist/university-content/notes/N-para-501-review-compliance.md

@@ -0,0 +1,72 @@
+---
+id: N-para-501-review-compliance
+title: Automated Review Pipeline & Compliance Checking
+type: note
+author: paradigm
+created: '2026-04-22'
+updated: '2026-04-22'
+tags:
+  - course
+  - para-501
+symbols: []
+difficulty: beginner
+estimatedMinutes: 2
+prerequisites: []
+category: paradigm-core
+origin: imported
+source: courses/para-501.json
+---
+
+## paradigm review
+
+The automated review pipeline uses a two-stage protocol:
+
+### Stage 1: Spec Compliance (always runs)
+
+- **Purpose coverage**: All touched symbols registered in .purpose files
+- **Portal gate compliance**: Routes declared in portal.yaml with gates
+- **Aspect anchors**: Anchor files still exist, no drift
+- **Broken references**: Parent symbols exist
+- **Route coverage**: New routes have portal.yaml entries
+
+### Stage 2: Code Quality (--deep only)
+
+- **Security**: eval() detection, hardcoded secrets
+- **Convention**: console.log usage (use Paradigm logger)
+- **Test coverage**: Gaps in test files
+
+### ReviewFinding Format
+
+Each finding has:
+- **type**: blocking (must fix), improvement (should fix), note (informational)
+- **category**: purpose-coverage, portal-compliance, aspect-anchors, security, convention
+- **message**: Human-readable description
+- **suggestion**: How to fix it
+
+### Usage
+
+```bash
+paradigm review              # Staged changes
+paradigm review --pr 123     # PR via gh CLI
+paradigm review --ci         # Exit 1 on blocking
+paradigm review --deep       # Include code quality
+paradigm review --json       # JSON output
+```
+
+## Dynamic Tool Loading
+
+Tools are organized in three tiers:
+- **Core** (~15 tools): Always loaded (search, ripple, status, navigate, etc.)
+- **Feature**: Auto-detected from filesystem (lore → .paradigm/lore/, etc.)
+- **Advanced**: On-demand via `paradigm_tool_activate`
+
+## Response Format
+
+`response_format: 'concise'` on high-traffic tools strips secondary data:
+- paradigm_search: returns only { symbol, type }
+- paradigm_ripple: returns only { symbol, impact, summary }
+- paradigm_status: returns only { project, counts, total }
+
+## compliance-checker.ts
+
+Shared logic extracted from pm.ts postflight. Both `paradigm_pm_postflight` and `paradigm review` use the same compliance checks.

package/dist/university-content/notes/N-para-501-sentinel-deep-dive.md

@@ -0,0 +1,173 @@
+---
+id: N-para-501-sentinel-deep-dive
+title: Sentinel Deep Dive
+type: note
+author: paradigm
+created: '2026-04-22'
+updated: '2026-04-22'
+tags:
+  - course
+  - para-501
+  - symbolic-incident-records
+  - flow-position-tracking
+  - automatic-incident-grouping
+symbols: []
+difficulty: beginner
+estimatedMinutes: 6
+prerequisites: []
+category: paradigm-core
+origin: imported
+source: courses/para-501.json
+---
+
+## Beyond Stack Traces
+
+Traditional error tracking gives you a stack trace and a count. Paradigm Sentinel gives you *symbolic context* — which component failed, where in a flow it failed, what gate was being evaluated, and which known pattern matches the failure. This transforms incident response from "read the stack trace and hope" to "match against institutional knowledge and follow a resolution strategy."
+
+## Symbolic Incident Records
+
+When Sentinel records an incident, it captures both technical and symbolic context:
+
+```yaml
+id: INC-042
+timestamp: "2026-02-21T02:15:00Z"
+status: open
+error:
+  message: "Cannot read property 'id' of null"
+  stack: "at PaymentProcessor.processRefund (payment-processor.ts:142)"
+  type: TypeError
+symbols:
+  component: "#payment-processor"
+  flow: "$refund-flow"
+  gate: "^authenticated"
+flowPosition:
+  flowId: "$refund-flow"
+  expected: ["^authenticated", "^refund-eligible", "#process-refund", "!refund-completed"]
+  actual: ["^authenticated", "^refund-eligible", "#process-refund"]
+  missing: ["!refund-completed"]
+  failedAt: "#process-refund"
+environment: production
+```
+
+The `flowPosition` field is critical — it tells you exactly where in the defined flow the failure occurred. The refund flow expected 4 steps; only 3 completed. The failure happened at `#process-refund`, and the `!refund-completed` signal never fired. This immediately narrows the investigation to the refund processing logic.
+
+## Incident Grouping
+
+Sentinel automatically groups related incidents using symbolic similarity. When two incidents share the same component, flow, and error pattern, they form a group. The grouping algorithm uses a similarity threshold of 0.6 — incidents must share at least 60% of their symbolic context to cluster.
+
+An `IncidentGroup` tracks the common symbols, error patterns, occurrence count, first/last seen timestamps, and which environments are affected. If a group matches a known failure pattern, Sentinel attaches it as a `suggestedPattern`.
+
+## Failure Patterns
+
+Patterns are the institutional knowledge of your error handling. Each pattern defines matching criteria and a resolution strategy:
+
+```yaml
+id: payment-null-ref-001
+name: "Null reference in payment processing"
+pattern:
+  symbols:
+    component: "#payment-processor"
+  errorType: [TypeError]
+  errorContains: ["Cannot read property", "null"]
+resolution:
+  description: "Add null check before accessing refund object properties"
+  strategy: fix-code
+  priority: high
+  symbolsToModify: ["#payment-processor"]
+  filesLikelyInvolved: ["src/services/payment-processor.ts"]
+confidence:
+  score: 85
+  timesMatched: 12
+  timesResolved: 10
+  timesRecurred: 2
+```
+
+Six resolution strategies exist: `retry` (transient failure), `fallback` (use alternative path), `fix-data` (data issue), `fix-code` (bug), `ignore` (known harmless), and `escalate` (needs human decision). Pattern priority ranges from `low` through `medium` and `high` to `critical`.
+
+Patterns come from four sources: `manual` (team-created), `suggested` (Sentinel auto-generated from groups), `imported` (from another project), and `community` (shared patterns). Paradigm ships 26 seed patterns covering common failures like incomplete flows, gate bypasses, state race conditions, and unhandled signals.
+
+## The Triage Workflow
+
+Sentinel follows a defined lifecycle for incidents:
+
+1. **Record** — `paradigm_sentinel_record` creates the incident with error details, symbolic context, and optional flow position. The incident starts as `open`.
+
+2. **Triage** — `paradigm_sentinel_triage` lists incidents filtered by status, symbol, environment, or error text. The matcher automatically suggests patterns that fit each incident.
+
+3. **Investigate** — `paradigm_sentinel_show` with `includeTimeline: true` shows the full flow timeline — every gate passed, signal emitted, and state change leading up to the failure. With `includeSimilar: true`, it surfaces related incidents that may share a root cause.
+
+4. **Resolve** — `paradigm_sentinel_resolve` closes the incident with a resolution: which pattern applied (if any), the fix commit hash, PR URL, and notes. Resolved incidents feed back into pattern confidence scores.
+
+5. **Pattern** — `paradigm_sentinel_add_pattern` creates new patterns from resolved incidents. When you fix a novel failure, capture the fix as a pattern so the next occurrence resolves faster.
+
+The sequence is: **record → triage → show → resolve → add pattern**. This cycle builds institutional knowledge with every incident.
+
+## Stats and Health Metrics
+
+`paradigm_sentinel_stats` provides operational intelligence for a given time period: total incidents, open vs resolved counts, incidents by environment and day, pattern effectiveness (which patterns resolve most incidents vs which recur), symbol hotspots (components with the highest incident rates), and resolution metrics (average time to resolve, pattern vs manual resolution rates).
+
+The `symbolHealth` view shows per-symbol incident history — use it to identify which components need hardening or refactoring.
+
+## Logger Transports
+
+Sentinel integrates with the Paradigm logger through a transport layer. The `LogTransport` interface defines a simple contract: a transport receives structured log entries and delivers them somewhere — a file, a remote API, a database, or Sentinel's ingestion endpoint.
+
+```typescript
+interface LogTransport {
+  name: string;
+  send(entry: LogEntry): void | Promise<void>;
+}
+```
+
+The logger supports multiple transports simultaneously via `addTransport(transport)` and `removeTransport(name)`. By default, logs go to the console. Adding a `SentinelTransport` sends them to Sentinel's server as well, without changing any of your existing logging calls.
+
+## The SentinelTransport Bridge
+
+Connecting the Paradigm logger to Sentinel is a one-liner:
+
+```typescript
+import { enableSentinel } from '@a-company/sentinel';
+
+enableSentinel({ endpoint: 'http://localhost:3001' });
+```
+
+This call creates a `SentinelTransport` instance and registers it with the logger via `addTransport`. From that point forward, every `log.component(...)`, `log.gate(...)`, and `log.signal(...)` call is forwarded to Sentinel as a structured log entry. Error-level logs are automatically promoted to incident candidates.
+
+The beauty of this design is zero code changes to your application. Your existing logger calls remain unchanged — the transport layer silently bridges them to Sentinel's observability pipeline.
+
+## Metrics API
+
+Sentinel's server exposes a metrics API for recording and querying application metrics:
+
+**POST /api/metrics** — Record a metric data point. Supports three metric types:
+- `counter` — Monotonically increasing values (e.g., request count, error count)
+- `gauge` — Point-in-time values that can go up or down (e.g., active connections, queue depth)
+- `histogram` — Distribution of values over time (e.g., response latency, payload size)
+
+```json
+{
+  "name": "api.requests.total",
+  "type": "counter",
+  "value": 1,
+  "labels": { "method": "POST", "route": "/api/payments" },
+  "timestamp": "2026-02-21T14:30:00Z"
+}
+```
+
+**GET /api/metrics** — Query metrics with optional filters by name, type, labels, and time range. Returns aggregated data suitable for dashboards and alerting.
+
+## Traces API
+
+Sentinel supports distributed tracing through span trees:
+
+**POST /api/traces** — Record a trace span. Each span has a `traceId`, `spanId`, optional `parentSpanId`, `operationName`, `startTime`, `endTime`, and `tags`. Spans with the same `traceId` form a tree — the root span has no parent, and child spans reference their parent via `parentSpanId`.
+
+**GET /api/traces** — Query traces by operation name, service, time range, or minimum duration. Returns full span trees with timing breakdowns.
+
+## Service Registry
+
+Sentinel maintains a live registry of services reporting data:
+
+**POST /api/services** — Register or update a service. Each service entry includes name, version, environment, health status, and last-seen timestamp.
+
+**GET /api/services** — List all registered services with their current health status and metadata. This provides a real-time view of what is running and where.

package/dist/university-content/notes/N-para-501-session-intelligence.md

@@ -0,0 +1,104 @@
+---
+id: N-para-501-session-intelligence
+title: Session Intelligence
+type: note
+author: paradigm
+created: '2026-04-22'
+updated: '2026-04-22'
+tags:
+  - course
+  - para-501
+  - four-checkpoint-phases
+  - breadcrumbs-auto-track-every
+  - paradigmsessionrecover-restores-last
+symbols: []
+difficulty: beginner
+estimatedMinutes: 4
+prerequisites: []
+category: paradigm-core
+origin: imported
+source: courses/para-501.json
+---
+
+## The Session Problem
+
+AI agent sessions are ephemeral. When a session ends — whether by completion, crash, context exhaustion, or human interruption — everything the agent knew vanishes. The next session starts blank, with no memory of what was explored, decided, or partially implemented. Session Intelligence solves this with checkpoints, breadcrumbs, and a global brain that persists knowledge across sessions and even across projects.
+
+## Session Checkpoints
+
+Checkpoints are deliberate snapshots saved at phase transitions. There are four phases:
+
+| Phase | When to Checkpoint | What to Capture |
+|---|---|---|
+| `planning` | After reading requirements, before coding | Plan, approach, key decisions |
+| `implementing` | After starting code changes | Modified files, symbols touched, decisions made |
+| `validating` | After implementation, before tests | All modified files, test plan |
+| `complete` | Task finished | Summary, final file list |
+
+Create a checkpoint with `paradigm_session_checkpoint`:
+
+```
+paradigm_session_checkpoint({
+  phase: "implementing",
+  context: "Adding JWT auth middleware — RS256 signing, httpOnly refresh tokens",
+  modifiedFiles: ["src/middleware/auth.ts", "src/handlers/refresh.ts"],
+  symbolsTouched: ["#auth-middleware", "^authenticated"],
+  decisions: ["RS256 over HS256 for public key verification"]
+})
+```
+
+Only `phase` and `context` are required — everything else is optional. The context field should be a concise 1-3 sentence summary of your current state of mind. Think of it as answering "if I were teleported into this session right now, what would I need to know?"
+
+Checkpoints are stored in `.paradigm/session-checkpoint.json` and auto-expire after 7 days.
+
+## Breadcrumb Tracking
+
+While checkpoints are deliberate, breadcrumbs are automatic. Every MCP tool call generates a breadcrumb recording the timestamp, tool name, symbol being modified (if applicable), and a human-readable summary. Breadcrumbs are stored in `.paradigm/session-breadcrumbs.json` with a maximum of 50 entries (auto-rotating — oldest dropped when full).
+
+Breadcrumbs capture the narrative of a session: "searched for payment symbols → checked ripple on #payment-service → read auth middleware → modified #auth-handler → created ^refund-eligible gate." This trail lets the next session understand not just what was done but the reasoning path.
+
+## Session Recovery
+
+Recovery is the payoff. Call `paradigm_session_recover` (or let it happen automatically — recovery data is surfaced on your first Paradigm tool call in a new session) to get:
+
+- **breadcrumbs** — The last session's tool call trail
+- **lastCheckpoint** — The most recent checkpoint with phase, context, and details
+- **symbolsModified** — All symbols that were changed
+- **recentActivity** — A human-readable summary of what happened
+
+This is crash recovery for AI agents. If a session dies at 87% context with half-finished auth middleware, the next session immediately knows: phase was `implementing`, auth middleware was being added, RS256 was chosen, these files were modified, and tests still need to be written.
+
+## The Global Brain
+
+Session Intelligence extends beyond individual projects through the Global Brain at `~/.paradigm/`. This user-level directory stores:
+
+- **Global wisdom** — Antipatterns and decisions that apply everywhere (e.g., "never use HS256 for JWT signing in production")
+- **Global habits** — Behavioral overrides that apply to all projects
+- **Cross-project practice events** — Compliance data aggregated across projects
+
+The distinction between project scope and global scope is important:
+
+| Scope | Location | Applies To | Example |
+|---|---|---|---|
+| Project | `.paradigm/` | This project only | "Use Redis for caching in this app" |
+| Global | `~/.paradigm/` | All projects | "Always check fragility before modifying critical symbols" |
+
+## Wisdom Promotion
+
+When a project-local wisdom entry proves universally valuable, promote it to global scope with `paradigm_wisdom_promote`. This copies the entry from `.paradigm/wisdom/` to `~/.paradigm/wisdom/`, making it available in every project.
+
+For example, if a team discovers that "always wrap Express v5 async middleware in try-catch" prevents errors across multiple projects, promoting this wisdom means every future project session gets this advice automatically when touching Express middleware.
+
+## Handoff Persistence
+
+When context usage exceeds 80-85%, `paradigm_session_health` recommends a handoff. `paradigm_handoff_prepare` creates a structured handoff document with: summary of work done, modified files, symbols touched, next steps, and open questions. This document is stored alongside session data so the receiving session can `paradigm_session_recover` and pick up exactly where the previous session left off.
+
+The handoff is not just a note — it is a contract between sessions. The outgoing session declares what was done and what remains. The incoming session validates against the actual file state and continues.
+
+## Best Practices
+
+- Checkpoint at every phase transition — the cost is ~100 tokens, the value is crash recovery
+- Write `context` as if briefing a stranger with no prior knowledge
+- Promote wisdom that survives 3+ projects to global scope
+- Use handoffs proactively at 80% context, not reactively at 95%
+- Let breadcrumbs accumulate naturally — don't try to manage them manually