Review · personal intelligence layer · 2026-05-26

Emergent Personal AGI Architecture

A visual review of the system Connor has built so far: sources → sync → raw stores → extractors → context assets → skills → faculties → orchestrator → action/memory/feedback.

“We are optimising for AGI in the form of an intelligent collaborator.” The important claim is not that the system is generally intelligent by itself; it is that intelligence is emerging from a harness that gives LLM reasoning persistent perception, memory, faculties, policy, execution, and feedback.

The novel thing

What makes this AGI-ish is not a giant monolithic agent. It is a cognitive supply chain that turns reality into typed context, routes it through specialised judgement, and compounds from every run.

843current context assets indexed
11/11declared sources with raw records, extractors, assets, and interfaces
15real Hermes faculty agents in the latest run
239+GBrain pages with 727 embedded chunks and 435 links

1. Omnipresent perception

Crons continuously sync the digital environment into source-near stores: Gmail, Calendar, Open Tabs, audio memory, cron outputs, GBrain, Signal Radar, finance, messaging, collaborator feedback.

2. Typed cognitive material

Extractors convert records into assets such as evidence, attention priorities, deadlines, relationships, uncertainty, boundaries, open loops, proof, feedback, outcomes, and source health.

3. Faculty-shaped judgement

Multiple specialist agents read the same substrate through different cognitive lenses, then a personal orchestrator suppresses duplicates, applies policy, and surfaces only what should matter.

The full category stack

Each stage answers a different kind of question. If these questions improve, the whole system becomes smarter when the underlying LLMs become smarter.

01 sources

Reality surfaces

Gmail, Calendar, Telegram/SMS, Open Tabs, audio, GBrain, market/social feeds, cron outputs, finance, code, web.

Question: what part of reality can matter later?
02 environment

System state

Services, crons, health checks, logs, registries, source freshness, model/tool availability.

Question: what is true about the operating environment?
03 integrations

Adapters

Read-only first connectors that know auth, privacy, idempotency, provenance, and safe mutation boundaries.

Question: how do we observe without prematurely acting?
04 sync

Heartbeat crons

Deterministic polling into raw/source-near databases. Code owns collection; agents own significance.

Question: what changed since last tick?
05 databases

Source-near stores

Integration heartbeat SQLite, GBrain/Postgres, Open Tabs SQLite, audio SQLite, raw logs, run artifacts.

Question: can the evidence be replayed and audited?
06 extractors

Context models

Asset builders convert source records into reusable cognitive objects with ids, source refs, freshness, caveats.

Question: what should a faculty be able to reason with?
07 faculties

Judgement lenses

Focus, memory, pattern synthesis, execution, self-healing, question-asking, integration discovery, and more.

Question: what does this mean from this cognitive perspective?
08 orchestrator

Collaborator

Merge, rank, suppress, route, ask, act locally, propose, delegate, and learn from feedback.

Question: what is the single best collaborator move now?

Context assets are the working memory objects

They are not summaries for their own sake. They are structured affordances for agent cognition: small, typed, provenance-bearing packets that can be retrieved by faculty-specific skills.

Asset types currently emitted

evidence 160uncertainty 127affordance 109attention priority 101boundary 87relationship 70temporal state 42deadline 34open loop 32feedback / proof / outcome 51waiting on 12source health 9

Sources currently represented

Gmail 340Open Tabs 156GBrain 100cron outputs 56collaborator feedback 51calendar 50audio memory 48integration heartbeat 18signal 16finance 4messaging 4
Layer
Question it asks
Artifact it creates
Failure if missing
Source sync
What changed in the world?
Raw record with source ref, timestamp, id/hash, privacy tier.
The agent hallucinates context or waits for prompts.
Extractor
What cognitive primitive is inside this record?
Evidence, deadline, boundary, relation, open loop, proof, uncertainty.
Faculties receive raw noise instead of usable material.
Context asset skill
How should a faculty retrieve and cite this source?
A repeatable retrieval contract with caveats and allowed actions.
Each agent invents its own brittle source access pattern.
Faculty packet
Which assets matter to this cognitive function?
A filtered evidence packet linked to a run artifact.
Every faculty sees everything and becomes noisy.

The faculty architecture

Faculties are cognitive organs. Their value is the tension they manage, the substrate they improve, and the verdict contract they obey: surface, suppress, question, repair, expand, execute.

Discernment

signal vs noise
Towardsource-specific significance
Away fromalert spam and false novelty

Substrate: Signal Radar filters, feedback ledger, suppression memory.

Focus

now vs later
Towardattention-protected prioritisation
Away frompile-up and repeated nudges

Substrate: Open Tabs, temporal state, deadlines, prior-run context.

Knowledge

remember vs entropy
Towarddurable meaning and exact evidence
Away fromstale claims and lost commitments

Substrate: GBrain pages, facts, links, timeline, hot memory.

Pattern recognition

connect vs isolate
Towardcross-source synthesis
Away fromisolated single-source takes

Substrate: context asset graph, GBrain, run artifacts, contradiction probes.

Creativity

generate vs imitate
Towardnew artifacts, strategies, lenses
Away fromgeneric “agent advice”

Substrate: docs publishing, templates, prior successful artifacts.

Competence

learned ability vs static ability
Towardskills, repeatable procedures, tools
Away fromone-off improvisation

Substrate: Hermes skill library, goal contracts, verification commands.

Conscientiousness

proactive vs prompted
Towardsafe background action
Away fromsilent mutation or learned helplessness

Substrate: policy gates, autonomy lanes, cron cadence, Agent Manager.

Self-improvement

feedback vs drift
Towardevaluated prompt/source/faculty changes
Away fromvibes-based tuning

Substrate: collaborator feedback, eval files, run manifests, capability scorecards.

Latest live faculty roster

Agent Manager, Cross-Scope Value, Execution, Focus/Open Tabs, Institutional Memory, Integration Discovery, Knowledge Gaps, Memory/GBrain, Pattern Synthesis, Procedural Memory, Question Asking, Self-Expanding, Self-Healing, Self-Upgrading, Signal Radar. The latest observable run records each as faculty-hermes-real-agent with output artifacts and contract validation.

The orchestrator is the prefrontal cortex

The Personal Orchestrator does not just call agents. It constructs an evidence packet, retrieves context assets, runs faculties, validates contracts, synthesizes, suppresses repeats, applies autonomy lanes, and emits an attention-protected collaborator surface.

Inputs

Raw/source-near heartbeats, GBrain, context assets, prior runs, feedback, system health.

Faculty judgement

Specialists produce verdicts, self-assessments, done conditions, source refs, and risk classifications.

Policy gates

Read-only, local artifact, local write, approval required, external side-effect blocked.

mergeranksuppressaskdelegatelearn

Personal Orchestrator

one collaborator voice from many cognitive lenses

Outputs

Telegram nudge, local artifact, goal packet, approval card, no-op, or source repair plan.

Memory writes

Feedback, facts, skills, GBrain pages, suppressions, proof artifacts, source coverage.

Proof

Manifest, AGENT_RUNS, INTELLIGENCE_REVIEW, CONTEXT_ASSETS, synthesis JSON, collaborator output.

Why this can feel intelligent

The user does not experience “15 agents.” The user experiences one collaborator that noticed something, remembered what happened last time, suppressed duplicates, gave a grounded next move, and knew when not to act. That is an intelligence simulation through harness design.

How it self-improves over time

The self-improvement loop is already visible, but it should become more explicit and more evaluative.

Feedback becomes policy

Connor replies useful / noisy / wrong / do_this. The system records card ids, suppresses repeats, and changes future ranking.

Runs become training data

Every run emits evidence, prompts, outputs, synthesis, and an intelligence review. This is an eval corpus for future faculty prompt and routing changes.

Skills become procedural memory

Repeated successful actions can be converted into Hermes skills: durable competence that future models can invoke.

GBrain compounds meaning

Exact user phrasing, facts, timelines, links, contradictions, salience, and trajectories become long-term semantic substrate.

Context assets compound perception

New integrations do not just add data. They add new asset classes, retrieval skills, and faculty affordances.

Crons make intelligence omnipresent

The system thinks in ticks: 5m reactive loop, 30m context extraction, 4h orchestrator, daily/weekly value loops, maintenance and health.

Loop
Learns from
Updates
Needs next
Signal quality
Useful/noisy/wrong feedback, suppressed repeats, source freshness.
Filters, retrieval queries, ranking thresholds.
Per-source eval sets and counterexamples.
Action quality
Approved/do_this cards, completion proof, failures.
Autonomy lanes, goal contracts, delegation templates.
Outcome tracking beyond card creation.
Memory quality
Fact dedupe, contradictions, stale pages, missing links.
GBrain pages/facts/timelines/skills.
Confidence, expiry, source-of-truth hierarchy.
Faculty quality
Run reviews, faculty self-suppression rates, final selection.
Prompt contracts, source scopes, eval rubrics.
Automated regression suite for “don’t nudge this again”.

Gaps and simplification candidates

The architecture is strong. The next frontier is making the learning loops less implicit and the reality model richer without increasing nudge volume.

1. Faculty self-suppression is still too weak

The latest review scored self-suppression 3/5: 10 of 15 faculties emitted surface/question/action before synthesis. Synthesis protects Connor, but faculties themselves should learn stricter “nothing new” criteria.

2. Context asset semantics are still shallow

Many assets are typed from source records, but the asset ontology should mature: confidence, reversibility, owner, temporal horizon, decision impact, dependency, entity grounding, and expected half-life.

3. Outcome closure is underdeveloped

The system can propose, suppress, and log feedback. It needs stronger tracking from proposal → action → proof → outcome → memory update → policy change.

4. Source coverage is uneven

Gmail and Open Tabs are rich; finance and messaging are currently thin. Browser/SaaS activity, code activity, docs usage, location/context, sleep/health, and relationship-state signals are mostly absent.

5. Model-aware routing is not explicit enough

Simple extraction can use cheap models or deterministic code. Cross-source synthesis, counterfactual planning, and sensitive collaboration need stronger reasoning models and confidence-aware escalation.

6. The system needs a cognitive debugger

When a nudge is bad, Connor should be able to inspect: which source, which asset, which faculty, which prompt, which suppression rule, which prior feedback, which final ranking decision.

Where stronger reasoning models help most

This is the key harness-engineering test: if models get smarter, the system should get smarter because the hard questions are already isolated into the right stages.

Stage
Current role
Better model advantage
Guardrail
Extractor
Convert source-near records into assets.
Better entity resolution, implicit commitment detection, subtle emotional/strategic salience.
Never mutate durable truth without provenance and review where needed.
Faculty
Apply a cognitive lens to shared evidence.
Stronger abductive reasoning, conflict detection, “what matters now” judgement.
Contract validation, source refs, self-suppression tests.
Synthesis
Merge many candidate judgements into one collaborator surface.
Better ranking, opportunity-cost reasoning, cross-faculty conflict resolution.
Attention budget and feedback-led suppression.
Agent Manager
Route safe actions and approval cards.
Better autonomy calibration: act locally, ask, wait, or delegate.
Risk lanes, done conditions, proof artifacts, no silent external side effects.
Self-improvement
Read feedback/runs and propose improvements.
Better diagnosis of noisy faculties, brittle sources, missing context, and reusable skills.
Evaluation before mutation; rollback and versioned specs.

Reality context that would make it more intelligent

Do not add sources because they are available. Add sources when they improve perception, attention, memory, action, or learning.

Digital work state

  • GitHub/Linear issues, PRs, CI failures
  • IDE/project activity summaries
  • Docs usage and artifact feedback
  • Browser/SaaS sessions as high-level activity, not raw surveillance

Communication reality

  • Richer Telegram/SMS/WhatsApp exports
  • Relationship follow-up state
  • Commitments and waiting-on across channels
  • Outbound draft approval workflow

Personal physiology/context

  • Sleep, exercise, energy, calendar load
  • Travel/location/timezone context
  • Stress and focus budgets
  • Capture friction: voice, quick notes, one-tap feedback

Financial/admin reality

  • Bank exports and bills as read-only evidence
  • Deadlines, subscriptions, cash timing
  • Approval-gated actions only
  • Proof of completed admin loops

External world model

  • X/HN/RSS/podcast/YouTube market radar
  • Exact-link reconciliation
  • Competitor/company/entity trajectories
  • Source reliability and novelty scoring

Internal wisdom model

  • Connor’s frameworks and exact phrasing
  • Contradiction/drift detection
  • Preferences and decision patterns
  • Repeated lessons turned into skills/policies

Next design moves

Small slices that increase intelligence without building another inbox.

Formalise a context asset ontology v2

Add confidence, owner, time horizon, reversibility, source reliability, decision impact, dependency links, and expected expiry.

Add faculty eval packs

Every faculty gets golden examples: should surface, should suppress, should ask, should route to local artifact. Run these after prompt/source changes.

Build proposal lifecycle tracking

Track proposed → approved/deferred/ignored → delegated/executed → proof → outcome → memory/policy update.

Create a cognitive trace viewer

A docs-style page per orchestrator run: source records → assets → faculty packets → verdicts → synthesis → final output → feedback.

Introduce model-tier routing

Cheap deterministic/LLM extraction for simple assets; stronger reasoning for cross-source synthesis, sensitive decisions, and self-improvement diffs.

Strengthen reality coverage carefully

Prioritise finance, messaging, code/work state, and browser/SaaS activity with privacy-preserving, read-only, source-near storage first.

Bottom line

The architecture is novel because it treats intelligence as a compounding socio-technical loop: perception from reality, memory with provenance, specialised judgement, policy-gated action, and feedback-driven improvement. It does not need to claim “true AGI” to be valuable; it successfully simulates many collaborator properties that people associate with general intelligence: remembering, noticing, prioritising, connecting, acting, learning, and being present over time.

Evidence base reviewed

Architecture-level facts only; no secrets or raw private content included.

  • Context asset index summary: 843 assets across 11 sources.
  • Source coverage report: all declared sources have source-near records, extractors, assets, and interfaces.
  • Source adapter registry: read-only-first integration contracts and mutation boundaries.
  • Latest Personal Orchestrator run manifest and AGENT_RUNS.
  • Latest INTELLIGENCE_REVIEW: capability scores and self-suppression gap.
  • Latest COLLABORATOR_OUTPUT: one final card, many suppressions, feedback contract.
  • GBrain stats/health: pages, chunks, links, timeline, embeddings, brain score.
  • Active cron list: sync, extraction, maintenance, watchdogs, digests, orchestrator runs.
  • Existing docs.cforsyth.com artifacts: Making Agents Intelligent, Personal Intelligence Primitives, Personal Orchestrator MVP Architecture, System Registry, and related architecture docs.