AEGIS
AEGIS OS · The Loop
Public / shareable guide for business and technical readers
Updated 2026-06-08
🔄 How It Works

The AEGIS Loop.
Vigilance that compounds.

Your AI team catches what you miss, stores what matters, and gets sharper with every session.

The core idea: most AI tools answer questions you remember to ask. AEGIS runs a continuous loop: it scans for what slipped, surfaces it with evidence, waits for your approval, remembers what you keep, and improves the next run overnight.

The five-stage vigilance cycle

1
🎯
CATCH
Radar scans CRM, email, tasks & memory
2
💡
SURFACE
Findings appear in your dashboard
3
APPROVE
You choose what's worth keeping
4
🧠
REMEMBER
Stored in your knowledge graph
5
⬆︎
IMPROVE
Skills evolve from your feedback
♾  Nightly & every new session — richer context, sharper catches

What happens at each stage

Expand any step. Each one has a plain-English explanation for context, and a technical breakdown for the engineers in the room.

01
🎯
CATCH — Radar scans everything
Your AI team sweeps every data source and flags what stands out

Think of this as your AI team running a morning sweep. They check the places work usually slips: CRM, email, tasks, calendar, and your knowledge graph. Then they produce a structured briefing: what is stalling, what went quiet, what follow-up is missing, and what risk is emerging. You did not have to ask.

Under the hood

The Radar aggregates a tenant-scoped snapshot from CRM, email, tasks, calendar, and Memoria. A tier-aware model resolver picks the allowed model rung for that tenant and intent, then returns a structured JSON envelope with five sections: top_opportunities, stale_opportunities, next_followups, content_gaps, risks_and_blockers. The runner is read-only and can be triggered manually or on a schedule.

02
💡
SURFACE — Findings land in your dashboard
Structured catches — nothing has been acted on yet

The results appear as a briefing in your AEGIS dashboard. Each catch has a title, a short summary, and a suggested next action. This is an information layer: AEGIS is showing you a picture, not making decisions. Nothing has changed in your CRM, inbox, or other systems.

Under the hood

The structured envelope is stored as a skill_runs record. Setup gaps such as disconnected integrations or empty sources are filtered before display so the dashboard only shows business signals.

03
APPROVE — You choose what matters
The human gate — nothing writes to memory without your OK

For each catch, you decide whether it belongs in your business's permanent knowledge. Click Accept and it goes into Memoria. Ignore it and it does not. This gate is non-negotiable: AEGIS will not decide on its own what is strategically important. That judgment stays with you.

Under the hood

Clicking Accept calls the approval flow for the Radar item. Items are blocked from acceptance if they are setup gaps, missing evidence, or lack a valid section classification. Only items that pass validation proceed to Memoria writeback.

04
🧠
REMEMBER — Stored in your knowledge graph
Persistent institutional memory that compounds with every session

What you approved does not disappear when you close the dashboard. It becomes part of your business's living knowledge graph: Memoria. Every agent can read from it, so the context you build now is available the next time the team briefs you.

Under the hood

Accepted items are written to kb_nodes with trust_level=human_approved. Nightly jobs apply confidence decay and crystallization so the graph stays current and useful.

05
⬆︎
IMPROVE — Skills evolve overnight
Your feedback rewrites underperforming AI procedures — automatically

Every time you rate a skill output with 👍 or 👎, you're teaching the system. AEGIS tracks which procedures work and which do not. Overnight, it rewrites the weaker ones and notifies you when a better version is active.

Under the hood

Every skill run writes a skill_runs row with a rating field. The nightly auto-learn job looks for underperformers, generates an improved prompt version from recent feedback, saves it to prompt_history, and applies it immediately.


How the automation actually triggers

AEGIS runs two categories of background jobs. One you configure; one runs automatically for all tenants.

System jobs — always running

JobRuns atWhat it does
job_syncer Every 60 s Picks up new tenant schedules from the database and registers them with APScheduler
job_nightly_memoria 02:00 AM Confidence decay on all kb_nodes + crystallization tier promotion
job_auto_learn 03:00 AM Scans skill_runs feedback, generates improved prompts for underperformers
job_nightly_pulse 02:00 AM Business Pulse CRM refresh — pipeline health metrics recalculated

Tenant schedules — you configure in Workflows

Set a cron expression (e.g. 0 8 * * 1-5 = weekdays 8am), choose a task prompt and an agent. Within 60 seconds, the job_syncer registers it. When it fires: APScheduler → execute_task() → openclaw → LiteLLM → agent response → result stored → Telegram notification (if enabled). If the LLM is temporarily unavailable (quota, timeout), a neutral delay notice is stored and the error is never surfaced to your Telegram. The schedule runs again automatically at its next interval.


What this looks like versus raw AI

Raw AI (one-shot) AEGIS Loop
Forgets everything after the session Memoria graph grows with every approved catch
You bring all context manually CRM, email, tasks auto-injected before every LLM call
Same outputs every time Skills improve from your feedback — every run is logged
You have to remember to ask Radar surfaces what you didn't ask about
No audit trail Every run, approval, prompt version, and decision logged
Platform owns your data Schema-per-tenant — your data never touches another account

Frequently asked questions

No. Nothing in the core loop writes to your knowledge graph, moves a lead, or sends a message without your explicit approval at stage 3. The APPROVE step is a contractual boundary. Scheduled tasks in Workflows execute the prompt you wrote — they do not make strategic decisions on your behalf.
CRM leads and contacts, email snippets from connected IMAP accounts (subject line, sender, date — never full email bodies), active tasks, calendar events if connected, and Memoria nodes already in your knowledge graph. The runner is fully read-only — it does not modify any data source.
Auto-learn requires at least 5 rated runs per skill before it evaluates evolution candidates. This prevents a single bad rating from triggering a rewrite. With regular daily use, most skills see their first evolution within 1–2 weeks. You can check skill rating history in the Skills → Intelligence tab.
Simply don't approve it. The item is never written to Memoria and has no effect. For best results, rate the Radar output with 👎 — this feeds the auto-learn cycle and teaches the Radar to surface better signals next time.
Yes. System jobs (Memoria decay, auto-learn, Business Pulse) and your configured Workflow schedules run continuously on the AEGIS server regardless of your session state. Results and task notifications reach you via Telegram as they complete.
The Opportunity Radar and Memoria writeback are available on all tiers (Free, Pro, Business, Enterprise). Skill auto-learn requires the auto_learn_mode setting enabled, available from Pro (€49/mo) upward. Model selection is governed by tier and intent.

Related guides

See the loop in action

8 agents. Persistent memory. Skills that improve every run. Free to start.