AI Memory vs Memori

AI Memory 65

Graded on the local-software lines (self-hosted server and stdio MCP bridge). Releases install from GitHub with install.sh or install.ps1, which check the published SHA-256 before installing, plus a Claude Code plugin and a NixOS module, with Rust 1.95 stated (20). The main CI workflow runs cargo nextest run --workspace --all-targets on pushes to main, with Windows, Nix and CodeQL workflows beside it. Run state couldn't be read on 4 November (api.github.com refused), so 15 of 25, from 0. Open crash and regression issues weren't read (github.com refused) (0). A Keep a Changelog file under semantic versioning, with 3.0.0 calling out the removed 1.x token commands and a migration and rollback guide (15). Version 3.3.3 (15). 20+15+0+15+15 = 65.

Memori 28

Hosted lines, since both modes call Memori's API. No status page found; status.memorilabs.ai did not answer and none is linked (0). No readable incident history (5). Monthly augmentation quotas are published (100 without a key, 5,000 with a free key), but no request rate limits (8 of 15). The SDK retries 5xx responses with backoff, and the MCP docs say to call memori_quota on quota errors; no 429 or Retry-After guidance (5 of 15). SLA only as a custom Enterprise term (0). No beta label on the cloud or MCP endpoint (10). Total 28.

AI Memory 90

25 MCP tools in crates/ai-memory-mcp/src/server.rs. Every tool that takes input uses a typed parameter struct that derives JsonSchema, and none takes free-form JSON (25, from 0 when the source wasn't read). Markdown guides in the repository, including INSTALL_FOR_AGENTS.md (10). Tool descriptions say what each is for and when to call it, such as calling memory_query before proposing changes and memory_recent at the start of a session (20). Typed inputs with documented fields (10 of 15). Setup and recovery examples (10 of 15). Releases and a dated changelog (15). 25+10+20+10+10+15 = 90.

Memori 68

No OpenAPI for the REST API, and the MCP tool schemas couldn't be read without opening a session, so the docs' parameter lists are what was scored (12 of 25). llms.txt with Markdown docs for both modes (10). Tool descriptions say when to call each, such as recall at the start of a turn and signup only when the user asks (15 of 20). Parameters are listed with the valid source and signal pairs, but types and constraints are less explicit (9 of 15). Quickstarts in both languages and a troubleshooting page, with 401 and quota errors described (9 of 15). Keep a Changelog with semantic versioning (13 of 15). Total 68.

AI Memory 65

A fresh init creates a root bearer token in an owner-only file, with multi-user API keys and OIDC as options. Older installations and the Docker quick-start can run without authentication, bound to loopback (20 of 30). Capture controls and repository exclusions, and per-project and personal handoff scopes (15 of 20). docs/security-boundaries.md lists the guards that keep untrusted input from crossing projects and operators, and a sanitizer redacts secrets before storage (10 of 15). Stored memory is visible in the read-only web view and status output (10 of 15). SECURITY.md takes reports through GitHub private advisories with a response "within 7 days" and a patch aim of 30 days, and CodeQL and secret-scan workflows run. No bug bounty or certification (10 of 20, from 0). 20+15+10+10+10 = 65.

Memori 45

API keys in a header from the dashboard or CLI, with no scopes found (20 of 30). No read-only mode or per-tool restrictions (4 of 20). Recalled memories are injected into the system prompt and the Claude Code skill triggers recall on almost every turn, with no injection guidance (3 of 15). A dashboard shows memories, retrieval and usage (8 of 15). SECURITY.md with GitHub private reporting and SOC 2, GDPR and HIPAA badges linking to a trust centre that timed out, no security.txt. Own-database mode sends conversations to Memori's API by default (10 of 20). Total 45.

AI Memory 50

v3.3.3 of 3 November 2026 (30). 68 tags from 6 August to 3 November, from v2.x through v3.3.3 (20). Issue and pull-request replies couldn't be read (github.com refused) (0). Not in the official MCP registry, and the repository has no server.json (0 of 15). Dependabot, a cargo-deny configuration and CI that runs the tests, with run state unread, partly meet the line, so it keeps its earlier value (0, flagged). 30+20+0+0+0 = 50.

Memori 33

SDK 3.3.6 on 27 May 2026, 163 days before the check (10). No release in the 90 days to 6 November (0). Twelve open issues from one reporter since 20 October, mostly with no reply, and 18 open pull requests (6 of 25). Python and TypeScript SDKs, last released in May 2026 (12 of 15). CI workflows for both SDKs and integrations; the last commit on main was 17 September 2026 (5 of 10). Total 33.

AI Memory 60

Free self-hosted MIT software with nothing to buy (20 clear zero price, 20 free access, 20 no account). No machine payment protocol (0). Optional cloud embedding or LLM providers bill separately. 0+20+20+20 = 60.

Memori 45

No x402, MPP or L402 (0 of 40). Plan prices are public but sales-led, from $60,000 a year (10). A free key with 5,000 augmentations a month, with no card mentioned (20). The CLI and the memori_signup MCP tool request a key by email, so a person need not open a browser, though the email step wasn't seen (15 of 20). Total 45.

AI Memory 60

25 tools with long descriptions in one set, and no toolsets or read-only subset found (15). Search takes scope and filters, and recent pages take a count (20). Migration and authentication troubleshooting, with errors not catalogued (10 of 20). Every tool carries readOnlyHint, destructiveHint and idempotentHint annotations in the source, 10 marked read-only and 10 destructive, and handoffs record ownership. No idempotency keys on writes. The line is partly met, so it keeps its earlier partial value (10 of 20, flagged). Client bridge defaults and setup commands, with one language (5 of 15). 15+20+10+10+5 = 60.

Memori 65

Seven MCP tools with short parameter lists (22 of 25). Recall filters by project, session, date range and source (14 of 20). 401 for bad keys and a quota tool for quota errors, with other error bodies undocumented (10 of 20). No idempotency guidance or tool annotations documented (4 of 20). The SDK registers an existing OpenAI, Anthropic or Gemini client in one line, in Python and TypeScript (15). Total 65.