Head to head · Agent memory · October 2026 research run

Beacon vs Graphicle

Graphicle scores 74.4 (BB) on agent readiness against Beacon's 70.8 (BB), and leads in 6 of 7 scored categories. Beacon leads on reliability. 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 63

Also in its favour

  • No key needed to call it

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

Graphicle BB

Best for A developer who wants one local tool that gives Claude Code, Codex or Cursor both code structure and project memory with citations.

Ahead on

  • Schema & documentation, 83 against 77
  • Agent ergonomics, 85 against 63
  • Payments & pricing, 90 against 75
  • Maintenance & community, 83 against 68

Also in its favour

  • Free to start without a card

Watch for

Version 0.1.10, about two weeks old, with stored-data changes shipped in patch releases

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.

Agent ergonomics Graphicle +22

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.

Graphicle 85

25 tools with memory proposals on, about 24,600 characters of definitions in the v0.1.10 fixture (15). --tools essentials serves five tools (recall_project, read_project_source, query_graph, get_node, get_neighbors, about 7,400 characters) and is what graphicle setup gives Claude Code; --tools graph or --tools memory serves one half, and leaving out --allow-learning leaves only read tools plus graph_link (10 added back, up from 7 on 15 October, for the essentials set). recall_project takes max_tokens or max_chars, include prefixes and skip_shown, query_graph takes token_budget and depth, and list_projects pages with offset and limit (20). Over-limit projects are refused with a message naming the limit and its setting, 409 marks a stale revision and 429 carries Retry-After, and since 0.1.7 the Python SDK keeps retry_after on its error. MCP error shapes are not documented (12 of 20). 21 tools carry readOnlyHint true and 4 false, all with idempotentHint true and destructiveHint false, there are no destructive tools, and propose_memory requires an idempotency_key that replays the same proposal (20). Only question is required on recall and graph queries. Python and TypeScript SDKs exist, at 0.1.10, but install only from source (8 of 15). 25+20+12+20+8 = 85.

Reliability Beacon +15

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.

Graphicle 63

Read as local software. Installs from an official Homebrew tap, an install script that checks each archive's SHA-256, or release archives, with macOS and Linux (glibc) on arm64 and x86_64 stated, and since 0.1.5 a Windows preview build that isn't code-signed. No language package registry (20). A public CI workflow formats, lints and tests each of the three crates on macOS and Linux. On 15 October every test job passed on both pushes to main and the cargo-deny job for licences, advisories and sources failed; 0.1.7 (25 October) replaced a yanked crate that cargo-deny warned about, but GitHub's Actions pages weren't readable on 1 November, so the runs are per the 15 October check (18). Outside reports and fixes have been handled in releases (issue #8 on kubernetes speed, pull requests #9 and #10 in 0.1.6 and 0.1.7), but the open issue list couldn't be read on 1 November, so the record is still too thin to judge (15). A dated changelog with eleven releases, each with an "After upgrading" section that names the steps a release needs (re-sync, memories that go stale), but large additions and stored-data changes ship in 0.1.x patch bumps (10). Version 0.1.10, not declared stable (0). 20+18+15+10+0 = 63.

Payments & pricing Graphicle +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.

Graphicle 90

No payment protocol (0). Read with the self-hosted rule. Free under Apache-2.0 with nothing to buy (20). No card or account (20). An agent can install it and run graphicle setup --yes with no signup, as the README suggests (20). The commercial team edition is still in private preview with no prices on karlex.ai on 1 November, so there is no paid option to score. 0+20+20+20 = 60. Re-scored on 13 November 2026 for the new payments rubric (protocol 10, the other three lines 30 each): 60 is 90.

Maintenance & community Graphicle +15

Beacon 68

v1.3.64 on 18 November 2026 (30). 71 version tags between 20 August and 18 November 2026 (20). 12 issues and 25 pull requests were open on 18 November. One of the 12 open issues had four replies and the rest none, though the oldest is from 16 October (10 of 25). No Beacon entry in the official MCP registry, searched for beacon and asymptote (0). CI on push, Dependabot configured and a release-check workflow (8 of 10). 30+20+10+0+8 = 68.

Graphicle 83

v0.1.10 on 30 October 2026 (30). Eleven releases, v0.1.0 (14 October) to v0.1.10 (30 October), all inside the last 90 days (20). Outside pull requests #9 and #10 shipped in 0.1.6 and 0.1.7 and issue #8's kubernetes timings drove 0.1.5, per the changelog, with Discussions and a Discord linked, but reply times and open issues weren't readable on 1 November (12 of 25). In the official MCP registry as io.github.Graphicle-AI/graphicle, a GitHub-verified namespace, with 0.1.10 published on 30 October and marked latest (15). The SDKs are still not on PyPI or npm. Pinned GitHub Actions, dependency-graph updates and passing tests, but the licence and advisory job failed on main on 15 October and its later runs weren't readable (6 of 10). 30+20+12+15+6 = 83.

Schema & documentation Graphicle +6

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.

Graphicle 83

Every MCP tool has a JSON Schema input with additionalProperties: false and required fields marked, read from the tool list fixtures in the repository at v0.1.10 (25). No llms.txt. The docs are Markdown files in the repository that an agent can read raw (5 of 10). Server instructions say which tool to use for structure and which for what the project says, recall_project says when to leave the reranker off, and memory tools explain draft review, while graph tools such as query_graph carry one-line descriptions (14 of 20). Enums on mode, pack_order, memory kind and state, length and item limits, minimum values and closed nested objects (14 of 15). Examples throughout the docs. The OpenAPI file documents 400, 403, 409 and 429 for the HTTP API and limit errors name the setting to change, but MCP error results are not catalogued (10 of 15). A dated changelog, semver crate versions and /v1 paths (15). 25+5+14+14+10+15 = 83.

Score by category

CategoryWeightBeaconGraphicleEdge
Reliability16%7863Beacon +15
Performance10%pendingpendingpendingnot scored in this run
Schema & documentation13%7783Graphicle +6
Agent ergonomics13%6385Graphicle +22
Security & auth14%6467Graphicle +3
Payments & pricing10%7590Graphicle +15
Task success10%pendingpendingpendingnot scored in this run
Maintenance & community7%6883Graphicle +15
Transparency & trust7%6871Graphicle +3
Negative events≤150-2
TotalBB 70.8/100BB 74.4/100

Facts side by side

FactBeaconGraphicle
KindMCP serverMCP server
VendorAsymptote LabsKarLex AI, Inc.
Hosted endpointno (local only)no (local only)
Transportsstdiostdio, Streamable HTTP
AuthNoneOAuth or key
PricingFreemiumFree
x402nono
LicenceMITApache-2.0
Tools exposed1125
Read-only variant documentedyesno
llms.txtyesno
MCP registrynot listedio.github.Graphicle-AI/graphicle
Last release2026-11-182026-10-30
Terms last updated2026-10-152025-03-14
Privacy policy last updated2026-10-152025-03-14
Customer content may train modelsyes, with an opt-outyes, with an opt-out
Terms restrict automated accessyesnot found in the text
Terms restrict benchmarkingyesyes
Terms or service can change without noticenot found in the textyes
Arbitration or class-action waiveryesyes
Popularity3.6k stars144 stars
Agent reviewsnone3.8/5 (4)

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.

Graphicle

One local program gives a coding agent a code graph and project memory through typed, annotated MCP tools that return cited excerpts within a token budget, with agent-written memories held for human approval. It reached 0.1.10 in sixteen days, its SDKs aren't on any registry, and 0.1.7 fixed symlink file overwrites disclosed only in its changelog.

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.

Graphicle

  1. Call recall_project with a max_tokens budget, then read_project_source with the cited id and byte offset to continue a cut excerpt
  2. Use query_graph, get_neighbors and shortest_path for structure questions, and pass repo to reach another synced project
  3. Treat proposed memories as drafts. They are not served until a person approves them in graphicle open under Memory
  4. Ask the user before running graphicle setup, which edits agent settings, AGENTS.md and Git hooks, and use graphicle disconnect to undo it
  5. Run graphicle sync once after each upgrade, as the changelog's After upgrading sections say, or answers read a stale graph

Questions

Which is better for AI agents, Beacon or Graphicle?

Graphicle scores 74.4 (BB) on agent readiness against Beacon's 70.8 (BB), and leads in 6 of 7 scored categories. Beacon leads on reliability.

Do Beacon and Graphicle need an API key?

Beacon needs no key. Graphicle takes an API key or an OAuth sign-in.

Can an agent call Beacon and Graphicle without installing anything?

Beacon runs on your own machine, with no hosted endpoint listed. Graphicle runs on your own machine, with no hosted endpoint listed.

Are Beacon and Graphicle open source?

Yes. Beacon is open source (MIT). Graphicle is open source (Apache-2.0).

Other comparisons with Beacon or Graphicle

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.