lucasrosati/claude-code-memory-setup
Coding · tested 2026-09-24 · by the Hlido desk, not the vendor
In short: A well-documented setup guide and toolchain for giving Claude Code persistent memory and codebase awareness via Obsidian + Graphify — high-value as a recipe, but it's a configuration playbook, not a product.
8 PASS · 2 FAIL of 10 public-surface claims
Quick answer
lucasrosati/claude-code-memory-setup scores 82/100 (STEADY) on Hlido’s independent, hands-on test (reviewed 2026-09-24). STEADY reflects a well-maintained, MIT-licensed, clearly documented and actively updated guide that solves recognised Claude Code problems. Pricing: Open source (free entry point documented).
This project packages a complete method for two real Claude Code pain points: amnesia between sessions and token waste from re-reading files. Its answer pairs Obsidian (as a Zettelkasten of "what was decided" — declarative memory) with Graphify (a codebase knowledge graph for "how the code is structured" — structural memory), plus a chat-import pipeline, and it headlines "up to 71.5x fewer tokens per session." As a guide it is genuinely useful and clearly written, bilingual (EN/PT-BR), MIT-licensed, and actively maintained, and it addresses problems every heavy Claude Code user recognises. The honest framing matters, though: this is a configuration playbook stitching third-party tools (Obsidian, Graphify) together, not a standalone application Hlido can run and measure. The eye-catching 71.5x figure is the author's, is best-case and workload-dependent, and was not benchmarked here. Its durability is also tied to the underlying tools staying compatible with Claude Code as that product evolves. For a developer willing to invest an afternoon in setup, the payoff is plausibly large; for someone wanting a one-click install, this isn't it.
Why STEADY
STEADY reflects a well-maintained, MIT-licensed, clearly documented and actively updated guide that solves recognised Claude Code problems. Not higher because it is a configuration playbook over third-party tools rather than a runnable product (limiting what Hlido can independently verify, medium confidence), and its headline token-saving claim is best-case, author-reported, and unmeasured here.
Public-surface checklist
- PASS Repo reachable (required) — GH API 200 for lucasrosati/claude-code-memory-setup
- PASS Readme present (required) — README length 22243
- PASS License present (required) — MIT
- PASS Install documented (required) — install/usage section found in README
- PASS Active 12mo (required) — last push 45d ago
- FAIL Releases present — no releases
- PASS Community traction — 852 stars
- FAIL Ci or tests — no .github/workflows
- PASS Recent commit 90d — last push 45d ago
- PASS Agent consumable — CLI install line in README
What we saw
1 screenshot captured by the Hlido engine during the reviewed run (run-7a2089a0fb0d5e76-github-com). Our own captures — not vendor marketing material.
What it does well
- Addresses two real, widely-felt Claude Code problems: cross-session amnesia and token-wasting re-reads
- Clear separation of concerns: Obsidian for decisions (declarative), Graphify for structure (structural)
- Thoroughly documented, bilingual (EN/PT-BR), with a step-by-step workflow and troubleshooting
- MIT-licensed and actively maintained (recent commits, community traction)
- Includes a chat-import pipeline to bootstrap memory from existing sessions
What it fails at
- It's a setup guide/recipe, not a standalone product — no single install, and it depends on third-party tools
- The "71.5x fewer tokens" headline is best-case, author-reported, and not benchmarked by Hlido
- Durability depends on Obsidian/Graphify staying compatible as Claude Code evolves
- Non-trivial setup effort; not suitable for users wanting a one-click solution
- Value is hard to verify independently because the outcome depends on the reader's own configuration
Best for
- Heavy Claude Code users willing to invest setup time for persistent memory and lower token use
- Developers who already use or are open to Obsidian for a second brain
- Teams standardising how they carry project context across AI-coding sessions
- PT-BR-speaking developers wanting first-class localised guidance
Not recommended for
- Users wanting a packaged, one-click product rather than a configuration playbook
- Anyone unwilling to adopt Obsidian and Graphify as dependencies
- Buyers who need the token-saving claim independently proven before investing setup time
- Environments that can't rely on third-party tools tracking Claude Code changes
Pricing & access
- ModelOpen source
- Free entry pointYes — a free tier or open-source edition is documented
Derived from Hlido-held evidence only (engine checklist + editorial text); quotes are verbatim from the scorecard; not vendor-supplied; re-derived daily. Verify current prices on the vendor's pricing page. Last verified 2026-07-16.
Related agents
Agent relevance
CLI
Agentic-Commerce Readiness 28/100 · SURFACE-ONLY
Independent readiness for agent delegation & transaction. How it’s scored · check live
A methodology for configuring Claude Code with Obsidian + Graphify for persistent memory and codebase awareness. Not a component agents call; rather, it shapes how a coding agent is set up. Value is in the workflow, not an API.
Agent-friendly score: 5/10
Score over time
The longitudinal record — every point is the score as published on that date. Raw series.
Evidence
- Public repository reachable with substantive README — source (2026-09-24) verified
- Product description: "Up to 71.5x fewer tokens per session on Claude Code with Obsidian + Graphify. Persistent memory, codebase knowledge grap" — source (2026-09-24) verified
- Open-source license present in repository — source (2026-09-24) verified
