# Memory Export Manifest Integrity Dashboard

Version: v3.104.0

Static reviewer-facing dashboard for checking .uai export manifests, llms exports, long-term source index, current file memory map, route/evidence manifests, AI discovery metadata, and latest archive snapshot for source-routed restricted-agent-safe completeness without manifest automation.

## Static review sections

### .uai/exports/manifest.json
- Expected location: `.uai/exports/manifest.json`
- Expected purpose: Machine-readable file-memory export inventory for restricted and advanced agents.
- Required current-version field: version
- Reviewer fields: pass, caution, no-op, human-review

### .uai/exports/llms.txt
- Expected location: `.uai/exports/llms.txt`
- Expected purpose: Compact restricted-agent safe memory summary.
- Required current-version field: latest package state
- Reviewer fields: pass, caution, no-op, human-review

### .uai/exports/llms-full.txt
- Expected location: `.uai/exports/llms-full.txt`
- Expected purpose: Full static memory export for capable agents that can handle larger context.
- Required current-version field: latest package state
- Reviewer fields: pass, caution, no-op, human-review

### /docs/long-term-source-document-index.md
- Expected location: `docs/long-term-source-document-index.md`
- Expected purpose: Long-term public source document index for research and memory sync inputs.
- Required current-version field: v3.104.0 entry
- Reviewer fields: pass, caution, no-op, human-review

### /docs/current-file-memory-map.md
- Expected location: `docs/current-file-memory-map.md`
- Expected purpose: Human-readable map of current docs, source research, .uai memory, manifests, exports, and validation coverage.
- Required current-version field: Updated for v3.104.0
- Reviewer fields: pass, caution, no-op, human-review

### assets/data/route-visual-manifest.json
- Expected location: `assets/data/route-visual-manifest.json`
- Expected purpose: Static route visual identity manifest.
- Required current-version field: version/themeVersion
- Reviewer fields: pass, caution, no-op, human-review

### assets/data/route-content-polish.json
- Expected location: `assets/data/route-content-polish.json`
- Expected purpose: Static route content polish and boundary manifest.
- Required current-version field: version/themeVersion
- Reviewer fields: pass, caution, no-op, human-review

### assets/data/public-evidence-packets.json
- Expected location: `assets/data/public-evidence-packets.json`
- Expected purpose: Public evidence packet manifest.
- Required current-version field: version/themeVersion
- Reviewer fields: pass, caution, no-op, human-review

### assets/data/ai-agent.json
- Expected location: `assets/data/ai-agent.json`
- Expected purpose: AI discovery metadata for safe machine readers.
- Required current-version field: themeVersion/currentRelease/version
- Reviewer fields: pass, caution, no-op, human-review

### latest .uai/archives snapshot
- Expected location: `.uai/archives/2026-06-04-teleodynamic-v3.104.0-memory-export-manifest-integrity-dashboard.md`
- Expected purpose: Immutable dated memory snapshot for this pass.
- Required current-version field: v3.104.0
- Reviewer fields: pass, caution, no-op, human-review

## Integrity comparison records

### v3.94.1 restricted-agent guidance memory sync
- Minimal Access Tier
- GET-Action Pattern
- code + url pointer responses
- no local execution

### v3.99.0 restricted-agent guide sync
- L0/L1 limited browser agents
- literal semantic keys
- idempotency and consent boundaries
- no secrets in query strings
- no-op escalation

### v3.103.0 file memory organization sweep
- current file memory map
- short-term memory
- progress
- file handoff
- export manifest
- llms exports
- archive snapshot

### v3.95.0 through v3.102.0 adoption archive chain
- final sign-off ledger
- publication receipt
- archive index
- completeness board
- decision ledger
- receipt verification matrix
- verification completion ledger

## Hard boundaries
- No runtime AI behavior
- No model training
- No live crawling
- No telemetry loops
- No agent execution
- No endpoint execution
- No credential validation
- No private-network probing
- No remote-site writes
- No webhooks
- No tunnel opening
- No screenshot automation
- No rollback automation
- No correction automation
- No sign-off automation
- No publication automation
- No archive automation
- No completeness automation
- No decision automation
- No receipt automation
- No verification automation
- No completion automation
- No memory automation
- No manifest automation
- No GET-Action execution
- No POST automation
- No safety certification
- No consciousness claim
- No biological-equivalence claim
- No medical/therapy claim
- No command-and-control claim over other domains