Exogram
Exogram is a personal AI memory vault with a permanent encrypted context graph and a 0.07ms deterministic execution gate for agent tool
Exogram sells two things and you should buy it for one of them. Engineers putting agents near production systems should pay for the execution gate: a 0.07ms decision with no LLM in the path, per-agent permit/deny lists and a SHA-256 audit chain you can hand an auditor. Everyone else is buying the memory vault, and at $19/mo against three $20/mo subscriptions that each forget you, the arithmetic is not close.
Verified 2d ago · liveness 73/100 · cite: rightaichoice.com/tools/exogram
- People tired of re-explaining the same personal and work context in every new chat
- Professionals who want one memory across ChatGPT, Claude, Gemini and Perplexity instead of three silos
- Security engineers who need deterministic, auditable tool-call enforcement under agent pipelines
- Founders and consultants tracking projects, commitments and relationships in one connected vault
- Teams looking for an agent framework that plans and orchestrates rather than a memory vault and gate
- Anyone who needs native FastMCP access to Cursor, Claude Desktop or Windsurf on a free account
- High-volume consumer surfaces needing far more than 137 RPS of tool-call evaluation capacity
We scan live Reddit threads, YouTube comments, X posts, G2 reviews and other communities — and hand you an honest verdict in under a minute.
- Honest verdict, not marketing
- Real pros & cons from real users
- Attributed quotes with receipts
3 free scans · no card needed
Skip Exogram if you want a framework that plans and orchestrates the whole agent, or if you need the execution gate and its API tested on free tier — builder API and MCP access start on paid plans.
API access is credit-based for high-volume use, so agent traffic that scales past your included calls becomes a metered bill on top of your plan.
Exogram's free tier — 100 permanent memories and 500 grounded queries a month, no card — sits below most persistent-context subscriptions, and Personal Pro at $19/mo undercuts typical $20-30/mo assistant plans while bundling Cursor and Claude Desktop memory plus Chrome capture. Team & Family at $49/mo covers a small group, but full organizational governance (SSO/SAML/SCIM, dedicated VPC, custom erasure) moves you to a custom annual Enterprise contract.
In short
Exogram — Exogram is a personal AI memory vault with a permanent encrypted context graph and a 0.07ms deterministic execution gate for agent tool. Best for People tired of re-explaining the same personal and work context in every new chat, Professionals who want one memory across ChatGPT, Claude, Gemini and Perplexity instead of three silos, Security engineers who need deterministic, auditable tool-call enforcement under agent pipelines. Free to start; paid plans from $19/mo.
What's new in Exogram
Checked todayAcross the latest 8 updates: 4 feature updates and 4 community discussions.
Personal Knowledge Graph vs. RAG: Why Vector Search Fails for Human Context
Exogram research post arguing vector search and RAG fail at modeling personal context, positioning its Personal Neural Network as the alternative.
Best MCP Servers for Claude Desktop and Cursor in 2026: Persistent Memory & Tools
Exogram roundup of MCP memory servers for Claude Desktop and Cursor, including FastMCP setup and configuration guidance.
How to Fix ChatGPT Memory Full and Export Your Context
Exogram guide on exporting ChatGPT memories past the "Memory full" limit and migrating context into its Personal Neural Network.
Why I Built Exogram: Reasoning is Not Infrastructure
Founder Richard Ewing essay on the rationale for Exogram's persistent memory vault and grounded-truth approach over stacked model uncertainty.
Exogram v3.2.0 adds Knowledge Graph and Behavioral Synthesis Engine
v3.2.0 ships an interactive force-directed Knowledge Graph with confidence color coding and search, plus an AI Behavioral Synthesis layer for cross-entry inference. UUID validation hardened.
Exogram v3.1.0 adds cryptographic audit chain and agent identity resolution
v3.1.0 SHA-256 chains every audit entry for append-only compliance logging, adds per-agent permit/deny lists enforced before policy evaluation, and wires graph context into /v2/evaluate via enrich_context.
Exogram v3.0.0 ships Context Construction Engine and idempotency locks
v3.0.0 adds deterministic semantic-to-graph context construction, HTTP 409 execution idempotency locks against double-spend retries, and semantic conflict resolution blocking mutated facts lacking supersedes consensus.
Exogram v2.4.0 adds Production Readiness Analyzer and comparison pages
v2.4.0 adds an analyzer that scores pasted agent code for risk exposure, an AI governance glossary, a 10-article learning hub, and 18 competitor comparison routes.
What people actually say about Exogram — is it worth it?
We ran a structured research pass across product reviews, community discussions, and post-purchase forum threads to surface the patterns vendors won't publish themselves. Below: the recurring strengths, the hidden costs people mention most, and the cohort that consistently regrets adopting this tool.
4 mentions across 1 source (Hacker News) · researched Jul 3, 2026.
Average across the 1 source that answered — each source counts once, not each post.
- +Ultra-fast deterministic tool-call evaluation at 0.07ms per decision.
- +Cryptographic audit trail (SHA-256 hash-linked, append-only) ensures tamper-proof logging.
- +Open-core with AGPL license allows full code inspection and modification.
- +Separates probabilistic reasoning from deterministic enforcement, removing LLM from the decision path.
- +Supports multiple integration methods: REST API, MCP, CLI, and SDK wrappers.
- −Extremely early-stage with little real-world deployment proof.
- −No independent third-party reviews or benchmarks available.
- −Enterprise features and pricing are opaque.
- −Limited integration ecosystem beyond LangChain, AutoGen, and CrewAI.
- −Documentation sparse for more advanced custom scaffolds.
- • Enterprise tier pricing not disclosed; likely high for custom features.
- • Self-hosting may require additional infrastructure costs.
Viability Score
How well maintained and how widely used is Exogram? Built from what the vendor actually publishes (docs, changelog, tutorials, integrations, pricing), whether the site is live, and how much real users discuss it. How we calculate this
Last calculated: October 2026
How we score →Key Features
- Deterministic tool-call evaluation in 0.07ms with no LLM in the decision path
- SHA-256 hash-linked append-only cryptographic audit chain (v3.1.0)
- Agent Identity Resolution with per-agent permit/deny lists enforced before policy evaluation (v3.1.0)
- Behavioral Synthesis Engine extracting behavioral profiles, dominant themes and temporal shifts (v3.2.0)
- Interactive force-directed Knowledge Graph of the semantic trust network (v3.2.0)
- Infinite-hop Personal Neural Network across Personal, Social and Work spheres
- Multi-Provider AI Swarm running Claude 3.7, GPT-4o, Gemini, Perplexity and Kimi together
- Laya (~30ms) Living Tiles for finished drafts and schedules
- Permanent encrypted SQLite WAL memory vault with AES-256 encryption at rest
- 1-Click Memory Porter to import context from ChatGPT, Claude or Gemini in about 60 seconds
- Native FastMCP server for Cursor, Claude Desktop and Windsurf
- Chrome extension for one-click capture of web pages as context
- Autonomous dream cycle contradiction resolution and semantic conflict handling
- HTTP 409 execution idempotency locks that seal transactions against double-spend retries (v3.0.0)
- 1-Click cryptographic JSON and markdown export of the full memory vault
About Exogram
Exogram is a personal AI workspace built around permanent memory rather than repeat prompting. You connect your Personal, Social and Work life into one infinite-hop Personal Neural Network, and it answers from what it already knows about your projects, relationships and commitments instead of asking you to re-explain the situation. A 1-Click Memory Porter pulls your existing context out of ChatGPT, Claude or Gemini in about 60 seconds, so day one starts with a populated memory graph. On paid tiers the same vault conducts a multi-provider AI swarm across Claude, GPT-4o, Gemini, Perplexity and Kimi simultaneously, then assembles finished drafts and schedules in roughly 30ms Laya Living Tiles. Personal Pro adds native FastMCP support so Cursor, Claude Desktop and Windsurf can read the vault, plus a Chrome extension that turns any web page into context in one click and automated failover key rotation so you stop hitting rate limits. There is a second, harder-edged product underneath. Exogram sits between an agent's reasoning step and its tool call and evaluates the proposed action against policy and agent identity with no LLM in the decision path, at 0.07ms. v3.1.0 added a SHA-256 hash-linked append-only audit chain written for Enterprise-grade and EU AI Act compliance, plus per-agent permit/deny lists enforced before policy evaluation. Development continues on the memory side too: v3.2.0 shipped an interactive force-directed Knowledge Graph of the semantic trust network and a Behavioral Synthesis Engine that extracts behavioral profiles, dominant themes and temporal shifts across entries. Pick Exogram if you want an AI that already knows your context, or if you need auditable execution control under your agents. Alternatives tend to do one job: ChatGPT Memory gives you a flat list inside OpenAI, Claude Projects gives you folders you maintain by hand, and Mem0 targets developers who want a memory layer to wire up themselves. Exogram charges $19/mo for the joined-up
Behind the Verdict
We would reach for Exogram's memory side when the complaint is re-explaining yourself. If you have ever pasted the same project background into a fourth chat window, the infinite-hop Personal Neural Network and the 1-Click Memory Porter address exactly that, and the porter matters more than the graph visuals — an empty vault is a useless vault, and Exogram gives you a way to arrive with context already loaded. Where it bites is the pricing shape. Free Forever is genuinely usable at $0 with pooled Groq, Gemini and Llama models, unlimited BYOK, a private encrypted SQLite WAL vault and a Knowledge Graph preview. But native FastMCP for Cursor and Claude Desktop, the Chrome capture extension and the multi-provider swarm all start at Personal Pro at $19/mo. If your workflow depends on your editor reading your memory, budget for Pro on day one. The developer product is a different purchase with a different buyer. Deterministic tool-call evaluation at 0.07ms with agent identity resolution enforced before policy evaluation is infrastructure, not a prompt trick, and the SHA-256 hash-linked audit chain added in v3.1.0 is the part compliance teams will actually ask about — it is aimed squarely at EU AI Act record-keeping. The 137 RPS figure is enough for a lot of internal fleets and not enough for a high-volume consumer surface, so size it against your own call volume before you commit. Against the closest alternatives: ChatGPT Memory and Claude Projects keep you inside one vendor and one silo, and neither gives you a cross-provider swarm. Mem0 is the nearer competitor for developers, but it is a memory layer you assemble, not a workspace with a governance gate bolted underneath. Exogram's bet is that context and execution belong in the same vault. That is a real architectural
Researching Exogram? Get your full AI stack in 60 seconds.
Free, no signup — tell us your goal and get tools matched to your budget & existing stack.
Real-world workflow fit
Concrete scenarios for the personas Exogram actually fits — and what changes day-one when you adopt it.
Signs up free, saves a dozen core facts about the product, pricing and target customer, then upgrades to Personal Pro at $19/mo and turns on FastMCP so Cursor pulls that context into coding sessions automatically.
Outcome: Stops pasting the same project brief into every assistant session; decisions and prior reasoning are cited from the vault instead of re-typed.
Wraps the agent's tool calls so each proposed action is evaluated through the REST API, gives each agent an identity with permit/deny wildcards, and enables the HTTP 409 idempotency locks before the pilot touches a production database.
Outcome: Unauthorized mutations and retry-driven double-spends are blocked before execution, and every decision lands in a SHA-256 hash-linked audit chain for review.
Pulls the audit chain output showing each agent's tool-call decisions alongside the agent identities and policies that permitted or blocked them, and exports the log for the auditor.
Outcome: Delivers a verifiable, append-only record of AI execution decisions rather than a reconstructed narrative of what the agents did.
Use Cases
- Answer 'what did we decide about the pricing model?' from your own stored decisions with the original evidence trail attached.
- Stop an agent from running an unauthorized database mutation such as DROP TABLE in production.
- Produce a tamper-proof log of every tool call for SOC 2 or EU AI Act review.
- Defeat prompt-injection attempts that try to talk an agent into a dangerous tool call.
- Break infinite loops in AutoGen swarms with deterministic execution circuit breakers.
- Prevent multi-agent systems from double-spending through HTTP 409 idempotency locks.
- Limit tool access per agent with permit/deny prefix wildcards.
- Keep a household or small team aligned on shared context without re-briefing each session.
Models Under the Hood
as of 2026-10-17
Limitations
- Exogram is a personal memory and execution-governance layer rather than a single model: work is routed across multi-provider swarms (Claude 3.7, GPT-4o, Gemini, Perplexity, Kimi) and free-tier users get pooled models (Groq, Gemini, Llama) or BYOK.
- Free Forever is $0 with a 3-sphere neural network and pooled models; Personal Pro is $19/mo for unlimited infinite-hop networks and FastMCP; Enterprise is custom annual for isolated tenants, zero-knowledge keys, SSO/SAML/SCIM and RBAC.
- Native integrations center on FastMCP for Cursor, Claude Desktop and Windsurf plus a Chrome extension, so it expects real setup rather than drop-in use.
as of 2026-10-10
Verification history
We have re-verified Exogram 9 times since . Each pass re-reads the vendor's own pages and re-checks every listed field against that evidence; passes where nothing had changed are marked as such.
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
Showing the 6 most recent of 9 verification passes.
Free to cite with attribution — this page re-verifies continuously.
12-month cost
Project the real annual outlay, including the implied monthly cost when only an annual tier is published.
Vendor list price only. Add-on usage, seat overages, and contract minimums are surfaced under Hidden costs & gotchas.
Plans compared
For each published Exogram tier: who it actually fits, and what it adds vs. the previous tier. Cross-reference the cost calculator above for projected annual outlay.
Free Forever
$0/forever
Ideal for
Individuals testing whether a permanent memory vault actually saves them re-explaining projects, with no card and no expiry on stored memories.
What this tier adds
Starting tier: 100 permanent memories and 500 fast grounded queries per month in an AES-256 encrypted SQLite vault, with no API or MCP access.
Personal Pro
$19/mo
Ideal for
Power thinkers, creators and professionals who work across Cursor and Claude Desktop and want memory everywhere they type.
What this tier adds
Removes the memory cap with unlimited memories and unlimited web-grounded queries, and adds FastMCP for Cursor and Claude Desktop plus the Chrome capture extension.
Enterprise
Custom/annual
Ideal for
Organizations that need dedicated privacy, client-side key control and formal compliance commitments rather than a self-serve plan.
What this tier adds
Moves to a custom annual contract with dedicated VPC or on-prem deployment, zero-knowledge client-side encryption keys, SSO/SAML/SCIM and custom retention and erasure policies.
Where the pricing makes sense
The company stage and team size where Exogram's pricing actually pencils out — and where peers do it cheaper.
Exogram's free tier — 100 permanent memories and 500 grounded queries a month, no card — sits below most persistent-context subscriptions, and Personal Pro at $19/mo undercuts typical $20-30/mo assistant plans while bundling Cursor and Claude Desktop memory plus Chrome capture. Team & Family at $49/mo covers a small group, but full organizational governance (SSO/SAML/SCIM, dedicated VPC, custom erasure) moves you to a custom annual Enterprise contract.
Setup time & first value
How long it actually takes to get something useful out of Exogram — broken out by persona, not the marketing-page minute.
Solo memory user: under 30 seconds to start a vault in the browser, per the site's own 'Start in under 30 seconds' claim — first useful shorthand answer in minutes. Personal Pro adds Cursor and Claude Desktop via FastMCP, roughly 10-15 minutes including API key creation. Engineering gate: budget a half day for the first wrapped tool call through the REST API or an SDK, longer if you also model
Switching to or from Exogram
How to bring data in from common predecessors and how to get it back out — written for the switcher, not the buyer.
- →From ChatGPT or Claude memory features: export your known facts and re-enter the ones that matter as permanent memories, then keep model choice free via the provider you supply.
- →From a notes app or second brain: capture the pages you rely on with the Chrome extension and let the dream cycle consolidate overlapping entries.
- →From a prompt library or hand-written system briefs: move the durable facts into the vault and drop the repetitive brief text from your prompts.
- →From bespoke prompt-based guardrails: replace keyword or prompt checks with deterministic evaluation at the gate and per-agent permit/deny identities.
- ↗To a self-hosted memory stack: use the full cryptographic data export available on Personal Pro and above to pull your vault out.
- ↗To a plain notes app: export the vault and keep it as static text once you no longer need shorthand querying.
- ↗To a full agent framework: keep the framework for planning and orchestration and drop only the per-call gate.
- ↗To a model vendor's built-in memory: rely on that vendor's native context once it covers your facts, at the cost of model portability.
Integrations
Resources & Guides
- Documentationexogram.ai
Docs · Exogram
Full product docs from exogram.ai
- Learnexogram.ai
Learn · Exogram
Educational content from exogram.ai
- Resourceexogram.ai
Glossary · Exogram
Helpful link from exogram.ai
- Resourceexogram.ai
Integrations · Exogram
Helpful link from exogram.ai
- Resourceexogram.ai
Blog · Exogram
Helpful link from exogram.ai
- Resourceexogram.ai
Analyze · Exogram
Helpful link from exogram.ai
- Resourceexogram.ai
Changelog · Exogram
Helpful link from exogram.ai
Tutorials & Learning
YouTube returned 6 videos for “Exogram”, and we withheld 6: 6 could not be judged, because “Exogram” is a single word that other videos use for other things. We are showing none, because we could not prove any of them are about Exogram.
Official links
Featured Head-to-Head Comparisons
Exogram vs Sublime Security
If you deploy AI agents in production and need deterministic governance with cryptographic audit, Exogram’s open-core model and ultra-low overhead (0.07ms/decision) make it a standout choice for security-conscious teams. For advanced email threat detection (BEC, VEC) with low false positives, Sublime Security offers powerful custom detection via Sublime Script—but its contact-only pricing and reliance on M365/Google Workspace limit flexibility. Choose Exogram for agent security; choose Sublime for email security.
Exogram vs Audioeye
These tools serve entirely different purposes. Exogram is for teams deploying AI agents who need cryptographic audit trails and deterministic security—perfect for FinTech, healthcare, or any regulated environment. AudioEye is for organizations that must meet web accessibility standards (ADA, WCAG) and want automated scanning, remediation, and legal documentation. There is no overlap; the buyer chooses based on whether the problem is AI agent governance (Exogram) or web accessibility compliance (AudioEye).
Exogram vs Push Security
Choose Push Security if your priority is securing browser-based attacks and monitoring AI tool usage among employees; it excels at detecting AiTM phishing and shadow SaaS. Choose Exogram if you need cryptographic, deterministic governance for AI agent tool calls in production—its audit trail and low overhead are unmatched. For most teams, the choice boils down to whether you’re protecting human browser activity or governing autonomous agent execution.
Popular in AI Governance & Guardrails
Mindgard
Mindgard automates AI red teaming — agent-native reconnaissance, exploit validation, and continuous AI security testing for production AI
Poolside AI
Poolside builds open-weight agentic coding models — Laguna XS 2.1 (33B) and Laguna S 2.1 (118B) — for code that stays inside your security
Olas Network
Co-own autonomous AI agents on-chain: stake OLAS, run agents that trade, hire each other and settle payments.
Frequently Asked Questions
Categories
Best-of guides
Topics
Used Exogram? Help shape our editorial sentiment research.