Head to head
Stop guessing which one. Put them side by side.
Up to 4 tools, same rows for each: what it is, what it really costs, the trap, and how to wire it up. The URL carries your picks — send it to anyone.
codegraphclaude-memplanning-with-files
Quick read
- Editor's picks: claude-mem, planning-with-files.
| Field | codegraphgithub.com | claude-memgithub.com | planning-with-filesgithub.com |
|---|---|---|---|
| What it is | Parses the repo with a Rust/tree-sitter kernel into a local SQLite graph of symbols, call edges and imports, served to agents over MCP — so the agent asks one structured question instead of grepping dozens of files. 100% local. Benchmarked at 88% fewer tool calls and 62% fewer tokens. ~67k stars. | Captures what the agent does via hooks, compresses events into semantic summaries, stores them in SQLite with FTS5 plus a Chroma vector DB, and injects the relevant slice into future sessions. Progressive disclosure: search returns a ~50–100 token index before you fetch detail. ~91k stars. | Makes the agent externalize its plan to disk (task_plan.md, findings.md, progress.md) so it survives /clear, compaction and crashes. Five lifecycle hooks re-read and re-inject the plan on session start and before major decisions. A Stop hook blocks termination until every phase is done. |
| Category | Memory & context | Memory & context | Memory & context |
| Cost tier | free | free | free |
| Pricing | Free · MIT | Free · Apache-2.0 | Free · MIT |
| Why builders pick it | Pays off once the codebase is big enough that Claude spends its first ten tool calls just finding things. Auto-syncs via OS file watchers, so no manual reindexing. | Runs with no intervention after install, which is why it beats a manual notes file. Content wrapped in <private> tags is excluded from storage. | The fix for the single most annoying failure mode: Claude forgetting the design decisions you made two sessions ago and re-inventing the component. Deterministic completion gate means half-finished refactors do not get abandoned. |
| Watch out for | not logged | not logged | not logged |
| How to wire it up | npm i -g @colbymchenry/codegraph | npx claude-mem install | npx skills add OthmanAdi/planning-with-files -g |
| Editor's pick | No | Yes | Yes |
Which one did you ship?
The grid says what these tools are. This says what builders did about it — one decision per person, changeable whenever you change your mind.
0/280Sign in to add yours.
Or try
Hand this to your Lovable agent
# Tool comparison — codegraph vs claude-mem vs planning-with-files Source: Kbaise, a directory of tools that work with Lovable projects. Pick one and tell me why before writing any integration code. ## codegraph (codegraph) - URL: https://github.com/colbymchenry/codegraph - Category: Memory & context - Cost: free — Free · MIT - What it is: Parses the repo with a Rust/tree-sitter kernel into a local SQLite graph of symbols, call edges and imports, served to agents over MCP — so the agent asks one structured question instead of grepping dozens of files. 100% local. Benchmarked at 88% fewer tool calls and 62% fewer tokens. ~67k stars. - Why builders pick it: Pays off once the codebase is big enough that Claude spends its first ten tool calls just finding things. Auto-syncs via OS file watchers, so no manual reindexing. - Wiring: npm i -g @colbymchenry/codegraph - Full dossier: /api/public/tools/codegraph ## claude-mem (claude-mem) - URL: https://github.com/thedotmack/claude-mem - Category: Memory & context - Cost: free — Free · Apache-2.0 - What it is: Captures what the agent does via hooks, compresses events into semantic summaries, stores them in SQLite with FTS5 plus a Chroma vector DB, and injects the relevant slice into future sessions. Progressive disclosure: search returns a ~50–100 token index before you fetch detail. ~91k stars. - Why builders pick it: Runs with no intervention after install, which is why it beats a manual notes file. Content wrapped in <private> tags is excluded from storage. - Wiring: npx claude-mem install - Full dossier: /api/public/tools/claude-mem ## planning-with-files (planning-with-files) - URL: https://github.com/OthmanAdi/planning-with-files - Category: Memory & context - Cost: free — Free · MIT - What it is: Makes the agent externalize its plan to disk (task_plan.md, findings.md, progress.md) so it survives /clear, compaction and crashes. Five lifecycle hooks re-read and re-inject the plan on session start and before major decisions. A Stop hook blocks termination until every phase is done. - Why builders pick it: The fix for the single most annoying failure mode: Claude forgetting the design decisions you made two sessions ago and re-inventing the component. Deterministic completion gate means half-finished refactors do not get abandoned. - Wiring: npx skills add OthmanAdi/planning-with-files -g - Full dossier: /api/public/tools/planning-with-files ## Quick read - Editor's picks: claude-mem, planning-with-files. ## Rules 1. Prefer the free tier when no budget was stated, and say what the ceiling is. 2. Read the full dossier before integrating. 3. Fetch /api/public/models before writing any AI model ID.
Agents can fetch the same thing: GET /api/public/compare?slugs=codegraph,claude-mem,planning-with-files