Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add Avijit07x/claude-db --skill cdb-scangit clone --depth 1 https://github.com/Avijit07x/claude-dbWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/avijit07x/claude-db/cdb-scan)<a href="https://agentmods.dev/skills/avijit07x/claude-db/cdb-scan"><img src="https://agentmods.dev/badge/skills/avijit07x/claude-db/cdb-scan.svg" alt="Measured on agentmods" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00072 | $0.00993 |
| Opus 5 | $0.00036 | $0.00496 |
| Sonnet 5 | $0.00014 | $0.00199 |
| Haiku 4.5 | $0.00007 | $0.00099 |
Grade A, and why
cdb-scan scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 8d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 98 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Map this project into memory
Everything else in claude-db's memory is testimony: it records what happened, turn by turn, as it happened. A fresh install knows none of that, and stays useless for weeks while it fills up.
This fills the gap from the one source available on day one — the code itself, in two passes.
Pass 1 — the code graph
Run this first, from the project root:
claude-db scan
It parses every supported source file and stores each symbol and each relationship between them: what calls what, what imports what, what extends what. It is deterministic, costs no tokens, and takes seconds. Report the symbol and edge counts it prints.
Once it has run, find_usages answers structural questions from the graph:
mode: "usages"— what references this symbol, with the relation on each linemode: "explain"— that, plus what the symbol itself reachesmode: "path"withtarget— how two symbols connect
Prefer those over reading files when the question is about structure.
Pass 2 — the written profile
The graph records what the code is. It cannot say why the project is built
this way, so the sections below capture that in prose. What you write here is
your reading of the codebase, not a record of events, so it is tagged
inferred and must never be phrased as history.
Work through the sections below in order. For each one, look at the actual
files, then call the remember MCP tool once with the given key.
key is what makes this re-runnable: writing the same key again replaces that
note rather than adding a second copy. Use these exact keys, and always pass
tags: ["inferred"] and kind: "context".
Before you start, call search for profile in this project. If notes already
exist, read them first and update what changed rather than restating it — a
second run should be an edit, not a rewrite.
1. profile:stack
Languages, runtime and version floors, frameworks, package manager, database,
test runner, build tool. Read the manifest (package.json, pyproject.toml,
go.mod, Cargo.toml) and the lockfile rather than guessing. Name versions
where a version constrains what can be written.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 8d ago First seen · 98 lines · 72 tokens per session scan A 1e64efe50677
cdb-scan is a skill published in the GitHub repository Avijit07x/claude-db (222 stars, last pushed 8d ago), licensed Apache-2.0. It adds 72 tokens to every session and 993 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
evolving-status
View Self-Evolving Loop session status, history, and memory metrics.
autolearn
Use when a verified non-trivial fix lands or existing solution docs need refresh. Not for unverified fixes.
resume
Load a previous session from TokenMizer graph memory. Returns a compact context block (100-600 tokens) covering goals, completed work, decisions, open tasks, and files. Inject this as system context to continue exactly where you left off. Use when user says "resume", "continue from last time", "load my project", "what…
checkpoint
Save the current session to TokenMizer graph memory. Creates a persistent checkpoint with all tasks, decisions, files, and errors — resumable in any future session. Use when user says "save", "checkpoint", "remember this", "I'm done for today", or session is getting long.
consolidate-to-one-home
Fold a scattered fact into a single home and point the rest at it. Use when the same rule, constant, or definition is restated in several files, or the user says "consolidate this", "make this the single source of truth", or "ssotize this". The consolidation is proposed and approved before any edit lands. To…
session-log
USE WHEN ending a substantial working session — user says "thanks/tomorrow/wrap up", or session shipped 3+ commits with no log today, or user asks for a summary. Distills the conversation into docs/sessions/YYYY-MM-DD- .md (context, what was done, key decisions, open questions, next steps). Hard cap 80 lines.…