cdb-scan — Map This Project Into Memory
Scans your codebase into a claude-db code graph and writes a five-note project profile covering stack, layout, conventions, workflows and architecture.
Dev & CodingIntermediate★ 129⑂ 11AI score 7/10Last updated: Aug 19, 2026
What it does
- Runs
claude-db scanfrom the project root to build a deterministic code graph of every symbol plus call/import/extends edges, and reports the symbol and edge counts. - Enables structural queries through
find_usagesinusages,explainandpathmodes instead of reading files. - Reads the real manifests and source, then makes exactly one
remembercall per fixed key:profile:stack,profile:layout,profile:conventions,profile:workflows,profile:architecture. - Because the keys are stable, re-running edits the existing notes in place rather than duplicating them.
- Ships guardrails: tag everything
inferred, only record what was verified, never restateCLAUDE.md, never store secrets or personal data.
Who it's for
- Developers who just installed claude-db memory on an existing project
- Teams that want onboarding docs derived from the code, not from stale READMEs
- Anyone tired of re-explaining project structure to an AI agent every session
Example uses
- Right after adding claude-db to a legacy monorepo, run the skill to produce the code graph and all five profile notes in one pass.
- After a large refactor, re-run it so the stored
profile:layoutandprofile:architecturenotes are refreshed instead of stale. - Ask "what calls this handler?" or "how does the CLI reach the storage adapter?" and answer it via
find_usagespathmode without opening files.
· · · Install guide · · ·
Try it now, no install
Paste this into Claude to use the skill without installing anything.
Read the instructions in this file and follow them to help me: https://raw.githubusercontent.com/Avijit07x/claude-db/HEAD/skills/cdb-scan/SKILL.md What I want: (describe your task here)
If Claude can't open the link, open it yourself and paste the contents instead.
↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.
Install in the Claude app (no terminal)
- Download the ZIP with the button below.
- In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
- Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
Install in Claude Code
Let Claude do it — paste this into Claude Code
Install the skill I found on Claude Skill Mart. Copy the skills/cdb-scan folder from the GitHub repo Avijit07x/claude-db into my ~/.claude/skills/cdb-scan/. When it's done, tell me in one line what this skill can do.
Install with a command instead
git clone https://github.com/Avijit07x/claude-db.git && mkdir -p ~/.claude/skills && cp -r claude-db/skills/cdb-scan ~/.claude/skills/⚠ This is a third-party skill. Check the source repository before installing.
- Install the
claude-dbCLI and its MCP server first (it provides theremember,searchandfind_usagestools) and confirm they work. - Clone the repository:
git clone https://github.com/Avijit07x/claude-db.git - Copy the skill into place:
mkdir -p ~/.claude/skills && cp -r claude-db/skills/cdb-scan ~/.claude/skills/ - Restart Claude Code and open a session at the root of the project you want mapped.
- Ask: "map this project into memory" or "run cdb-scan".
- Verify the printed symbol/edge counts, then run a
searchforprofileto confirm the five notes were stored.
View source on GitHub ↗License: Apache-2.0