Head to head · Agent memory · October 2026 research run

Beacon vs TencentDB Agent Memory

Beacon scores 70.8 (BB) on agent readiness against TencentDB Agent Memory's 54.6 (C), and leads in 3 of 7 scored categories. TencentDB Agent Memory leads on agent ergonomics and payments & pricing. Both do agent memory.

Best memory layers for AI agents · All 383 memory comparisons

Which one, for what

Beacon BB

Best for Developers running several coding-agent harnesses who want one local record of sessions and a reviewed memory their agents can recall.

Ahead on

  • Reliability, 78 against 51
  • Schema & documentation, 77 against 61
  • Security & auth, 64 against 35

Also in its favour

  • Agent-ready, a grade of BB or better
  • No key needed to call it
  • Runs on your own machine
  • No incidents deducted, where TencentDB Agent Memory loses 6 points for them

Watch for

Beacon Cloud content can be used for research, model training and sold de-identified datasets unless the user opts out, per the terms

TencentDB Agent Memory C

Best for Small teams that run several coding agents on a host they control and want shared, permissioned memory, skills and code context injected through a proxy.

Ahead on

  • Agent ergonomics, 68 against 63
  • Payments & pricing, 90 against 75

Also in its favour

  • Free to start without a card

Watch for

The core's bearer gate is off by default, and the deploy notes say enabling it breaks the proxy's auth and session setup

Why the edge

Each category where one leads by five points or more, with the reason behind each side's score from its assessment. The numbers are in the table below; this is what they stand for.

Security & auth Beacon +29

Beacon 64

Over stdio the server runs with the user's own permissions and needs no secret. The loopback HTTP transport has no authentication, and the docs warn against binding it beyond 127.0.0.1 (20 of 30). All local tools are read-only, memory needs a person's approval before an agent sees it, and intel_check_installed sends possibly private names only when include_private_names is set, with a dry_run (16 of 20). Activity tools return recorded prompts and command output, and no guidance for an agent reading them was found, though Beacon's threat rules flag prompt injection in traces (5 of 15). Beacon itself is a local record of every agent action, and its log keeps MCP activity (10 of 15). SECURITY.md and a disclosure policy with security@asymptotelabs.ai, signed and notarised macOS packages, SHA-256 checksums and a SOC 2 Type II post title in the docs. No bug bounty found, no security.txt on beacon.sh, and the SOC 2 post sits on asymptotelabs.ai, which is closed to our reader (13 of 20). 20+16+5+10+13 = 64.

TencentDB Agent Memory 35

Read as self-hosted software with a network API. Per-user sk-mem- keys, several per user and revocable, and OAuth2 for panel login. Against that, Memory Core's bearer gate is off by default and the .env template says setting it breaks the proxy's auth and session setup, the data plane takes team, agent and user IDs from the caller and needs only x-tdai-service-id, the proxy's rate-limit and session admin routes have no authentication, /v3/instance/destroy rests on the same unset bearer, the deploy README's defaults include an admin key, and the start scripts publish every port on all interfaces (8 of 30). Private-by-default assets with team, restricted and agent visibility, ACLs by user, role and agent, and read-only Knowledge tools. Data-plane deletes and chat-memory/clear make no user-level check and nothing asks for confirmation. Asset reads were scoped to the caller only in 2.0.2, and Agent and Task reads only in an unreleased commit of 28 October (8 of 20). The proxy injects stored memory, skills, imported documents and code into the system prompt, and we found no guidance on prompt injection or memory poisoning. Personal skills are private until shared after review (3 of 15). Participation logs, memory-generation logs and the raw L0 record let the operator trace what was written, and per-call analytics to ClickHouse is available but off by default (9 of 15). CONTRIBUTING.md asks for vulnerability reports by email to agentmemory@tencent.com rather than public issues, with no acknowledgement time, so a disclosure route without a stated time (7). No SECURITY.md on any branch, no security.txt on www.tencent.com, and no bounty or audit named. The npm plugin's postinstall runs a bash script that rewrites OpenClaw's installed files, with output sent to /dev/null. Two access-control fixes in 2026 are under the negative events.

Reliability Beacon +27

Beacon 78

Local-software lines, as a CLI and stdio MCP server the owner runs. Installs from a Homebrew tap, a signed macOS .pkg and .deb and .rpm packages, with macOS and Linux on amd64 and arm64 stated (20). A public CI workflow runs on every push to main, with 792 Go test files and a Node 20, 22 and 24 matrix for the SDK, but run results couldn't be read, so 15 of 25. 12 issues were open on 18 November 2026, eleven of them opened from 1 to 17 November, a few small bugs among proposals (20 of 25). The CLI changelog is per version and names changed behaviour, but new capabilities ship in patch versions and nothing is marked as breaking (8 of 15). v1.0.0 on 12 July 2026 (15). 20+15+20+8+15 = 78.

TencentDB Agent Memory 51

Local-software reading, as software the owner runs. Official multi-architecture images (amd64 and arm64) on Docker Hub's agentmemory account, named in the changelog, with Node 22.16 or later stated for the plugin and Python 3.9 or later for the SDK. Setup still needs a git clone for the start scripts, which pull :latest (18 of 20). Vitest suites are declared in the package files. The one CI workflow, pr-ci.yml, runs on pull requests to the main branch (the OpenClaw plugin) and installs, packs and checks the manifest, package size and skill-queue isolation without running the test suites, and nothing runs on the default branch, now feat/server_team. GitHub's Actions pages weren't readable, so no result was read (8 of 25). GitHub issues weren't readable, so open crash reports weren't counted. The changelogs list fixes for empty recall, lost memories, duplicated history and a Knowledge service that failed to start on OpenTelemetry drift, and outside pull requests (#31, #59, #79, #85) were merged into the plugin (8 of 25). Keep a Changelog with SemVer declared and upgrade notes, such as backend switches not migrating data and a v2 to v3 migration tool, but three version streams disagree (image tags 1.0.x for releases 2.0.x, the npm plugin at 1.0.3 while main's changelog stops at 0.3.6) and ROADMAP.md still says 2.0.1-beta.1 is current on 4 November (9 of 15). Release 2.0.3, but the README heads the Hub 'Team Memory Beta' and the SDKs are beta-only (8 of 15).

Schema & documentation Beacon +16

Beacon 77

Every one of the 11 tools has a typed JSON Schema input, and the intel tools set additionalProperties false (25). https://docs.beacon.sh/llms.txt indexes Markdown copies of the docs (10). Activity and memory descriptions are one line on purpose only. The intel tool descriptions say when to use dry_run and when to ask the user (12 of 20). Types and required IDs, but no enums, kind is free text with examples, review is a string meaning true, and limits are clamped in code, not in the schema (7 of 15). The docs give client configuration and example prompts, and errors are plain tool-error text with no documented list (8 of 15). A dated CLI changelog with 163 version entries (15). 25+10+12+7+8+15 = 77.

TencentDB Agent Memory 61

API reading for a self-hosted HTTP service. The Knowledge service has an OpenAPI 3.0.3 file of 28 paths with no security scheme. Memory Core's 109 endpoints and the proxy's routes are described only in Markdown references, and the proxy's main LLM paths are outside them (10 of 25). No website and no llms.txt. The docs are Markdown in the repository and readable raw (5 of 10). Each Core endpoint states what it does, its limits and caveats, such as which deletes skip user checks, and Knowledge's tool list marks its tools read-only. The API references and the changelog are in Chinese; the README and install guide are in English (11 of 20). Zod-checked inputs with bounds (1 to 100 messages, 8,192 characters per message, limit up to 100) and enums for atom types, but the reference records that version comes back as a string from atomic/update and a number from query (10 of 15). Worked request and response examples per endpoint and error tables per module, with the three different error-message formats set out (12 of 15). /v3 paths with v1 and v2 routes listed as deprecated, and a dated public changelog for every release since 0.1.0 (13 of 15).

Payments & pricing TencentDB Agent Memory +15

Beacon 75

Self-hosted rule, with the price lines read for Beacon Cloud. No x402, MPP or L402 (0 of 10). Plan prices for Pro and Team are public on beacon.sh/pricing, with nothing priced per unit (15 of 30). The open-source CLI is free with no card (30). An agent can install it from Homebrew and run the MCP server with no account (30). 0+15+30+30 = 75.

TencentDB Agent Memory 90

Read with the self-hosted rule. MIT software with nothing to buy from Tencent, so 20 for pricing, 20 for a free start and 20 for use without an account. No x402, MPP or L402 (0). Memory extraction, wiki and code-graph building and the proxied model are paid to whichever LLM provider the owner configures. Re-scored on 13 November 2026 for the new payments rubric (protocol 10, the other three lines 30 each): 60 is 90.

Agent ergonomics TencentDB Agent Memory +5

Beacon 63

11 tools, all compact, so the 11 to 30 band (15). Activity search takes limit, since, until and 15 substring filters. Memory search caps at 20 results, get_memory_context at 5, and memory bodies are cut to 1,200 characters (18 of 20). Errors are short text such as "id is required" or "memory not found" and name the wrong argument, without codes (12 of 20). Every tool reads and none writes, so retries are safe, but no tool carries readOnlyHint or other annotations (8 of 20). Only the ID tools need an argument, and the server starts with one command. One Go binary, plus a JavaScript SDK for tracing (10 of 15). 15+18+12+8+10 = 63.

TencentDB Agent Memory 68

Read with the API lines. limit defaults of 5 or 20 up to 100, recall.maxResults 5, and character budgets and a token ratio for what the proxy injects, but no field selection, and the proxy adds memory to every request's system prompt whether or not the agent asks (16 of 25). limit and offset on list and query routes, {items, total} on meta lists, filters by session, time range, atom type and team, agent, user and task IDs (18 of 20). HTTP codes with documented meanings (400, 401, 403, 404, 409, 422, 429, 503), but three message formats by module, /v3/meta/auth/verify answering 200 with valid: false for a bad key, and tools/call failing as HTTP 500 inside a code: 0 envelope (12 of 20). Wiki and code-graph creation and knowledge upserts are idempotent by name or repository, while /v3/session/add is documented as not idempotent and conversation/add takes no idempotency key (10 of 20). Zero-configuration defaults and an interactive start script, official TypeScript and Python SDKs in beta, and agents join by changing a base URL. Header preselection for Hermes and OpenClaw needs a task ID made in the panel first (12 of 15).

Score by category

CategoryWeightBeaconTencentDB Agent MemoryEdge
Reliability16%7851Beacon +27
Performance10%pendingpendingpendingnot scored in this run
Schema & documentation13%7761Beacon +16
Agent ergonomics13%6368TencentDB Agent Memory +5
Security & auth14%6435Beacon +29
Payments & pricing10%7590TencentDB Agent Memory +15
Task success10%pendingpendingpendingnot scored in this run
Maintenance & community7%6870TencentDB Agent Memory +2
Transparency & trust7%6868even
Negative events≤150-6
TotalBB 70.8/100C 54.6/100

Facts side by side

FactBeaconTencentDB Agent Memory
KindMCP serverHTTP API
VendorAsymptote LabsTencent Cloud
Hosted endpointno (local only)no (local only)
TransportsstdioHTTP
AuthNoneAPI key
PricingFreemiumFree
x402nono
LicenceMITMIT
Tools exposed11none
Read-only variant documentedyesno
llms.txtyesno
Last release2026-11-182026-10-26
Terms last updated2026-10-15no document linked
Privacy policy last updated2026-10-15no document linked
Customer content may train modelsyes, with an opt-outnot read yet
Terms restrict automated accessyesnot read yet
Terms restrict benchmarkingyesnot read yet
Terms or service can change without noticenot found in the textnot read yet
Arbitration or class-action waiveryesnot read yet
Popularity3.6k stars2.2k npm/wk

How the scores have moved

Verdicts

Beacon

Beacon records sessions from more than 20 coding-agent harnesses on the machine, under MIT, and its stdio MCP server returns only human-approved project memory. Releases ship almost daily. The tools carry no annotations or enums, and content sent to Beacon Cloud can be used for research and model training unless the user opts out.

TencentDB Agent Memory

A self-hosted team memory stack with layered recall, per-asset visibility and ACLs, documented HTTP APIs and frequent releases, 2.0.1 to 2.0.3 since August. The default deployment publishes its ports on all interfaces with the core's bearer gate off, the proxy's admin routes have no authentication, and two 2026 access-control fixes came without advisories.

Before you call either

Beacon

  1. Start the server with beacon mcp serve --transport stdio. Keep --transport http on 127.0.0.1, because the loopback endpoint takes no credential.
  2. Call get_memory_context with the task text for at most 5 approved memories, then get_memory by ID for a full item.
  3. Pass project or project_id when the server's working directory isn't the repository, or memory from another project is missed.
  4. Run intel_check_installed with dry_run: true first. Without it the tool sends package and skill names to intel.beacon.sh.
  5. Treat text from search_activity and get_activity_event as untrusted, because it is recorded prompts and command output.
  6. Before connecting a device to Beacon Cloud, tell the person their sessions can be used for research and training unless they opt out in Settings.

TencentDB Agent Memory

  1. Set MEMORY_CORE_GATEWAY_API_KEY, KNOWLEDGE_SERVICE_KEY and a non-default admin key before exposing ports 8420, 8424, 8096 or 8125 beyond the host
  2. In headless runs send x-team-id, x-agent-id and x-task-id headers to the proxy. Without all three it falls back to an interactive picker and skips memory
  3. Expect every proxied turn to be stored. The proxy records upstream messages and model replies to Memory Core, and injected memory lands in your system prompt
  4. Check data.valid on /v3/meta/auth/verify and data.isError on /v3/tools/call. Both can fail inside a success envelope
  5. Retry /v3/session/add only on 500, 503 or a network error. The API reference says it isn't idempotent and writes a new object each call
  6. Build from the default branch past 2.0.3, or wait for the next release, before giving several users one Memory Core. Agent and Task reads were unscoped until 28 October

Questions

Which is better for AI agents, Beacon or TencentDB Agent Memory?

Beacon scores 70.8 (BB) on agent readiness against TencentDB Agent Memory's 54.6 (C), and leads in 3 of 7 scored categories. TencentDB Agent Memory leads on agent ergonomics and payments & pricing.

Do Beacon and TencentDB Agent Memory need an API key?

Beacon needs no key. TencentDB Agent Memory needs an API key.

Can an agent call Beacon and TencentDB Agent Memory without installing anything?

Beacon runs on your own machine, with no hosted endpoint listed. No hosted endpoint is listed for TencentDB Agent Memory.

Are Beacon and TencentDB Agent Memory open source?

Yes. Beacon is open source (MIT). TencentDB Agent Memory is open source (MIT).

Other comparisons with Beacon or TencentDB Agent Memory

Machine-readable

For companies

Do agents find, use and choose your tools?

An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.