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 agentmods add commands/geckse/markdown-vdb/shardsgit clone --depth 1 https://github.com/geckse/markdown-vdbWhat 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 | $0.00012 | $0.01044 |
| Opus 5 | $0.00006 | $0.00522 |
| Sonnet 5 | $0.00002 | $0.00209 |
| Haiku 4.5 | $0.00001 | $0.00104 |
Grade A, and why
shards 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 3d 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 — 141 lines — stays where its author put it; the contents beside it link to each section on GitHub.
mdvdb shards
Shards give a collection-relative folder a stable name for humans, scripts, and agents. A Shard is a recursive logical lens over one Collection: it reuses the Collection's watcher, link graph, full-text index, and vector index. It is not a nested index or an access boundary.
Configuration
Definitions live only in the project's .markdownvdb/config.yaml:
shards:
research:
name: Research
path: work/research
topics:
- name: Methods
description: Research methods and experiments
seeds: [methodology, experiment]
threshold: 0.35
papers:
name: Papers
path: work/research/papers
papers is displayed beneath research because its folder is contained by the Research folder.
The relationship is inferred; no parent field is stored.
When --name is omitted on add/create, the CLI derives a title-cased display name from the
ID (research-papers becomes Research Papers).
Management
mdvdb shards list
mdvdb shards get research
mdvdb shards add research --name Research --path work/research
mdvdb shards create drafts --name Drafts --path work/drafts --create-dir
mdvdb shards update research --name "Research Library"
mdvdb shards update research --path archive/research
mdvdb shards retarget work/research archive/research
mdvdb shards remove research
mdvdb shards delete research
remove and delete never delete the folder or any Markdown files. Missing folders remain in
list with exists: false, which makes external renames repairable.
Using a Shard
The following commands accept --shard <ID>:
mdvdb search "vector databases" --shard research --json
mdvdb tree --shard research --json
mdvdb info --shard research --json
mdvdb schema --shard research --json
mdvdb collection --shard research --recursive --json
mdvdb graph --shard research --json
mdvdb modules run formula --shard research --json
mdvdb modules status formula --shard research --json
Except for graph projection described below, --shard and the command's path selector are
mutually exclusive. Results keep their collection-root-relative paths and document identities.
get, links, and backlinks continue to take explicit collection-root-relative document paths
and therefore work across Shards.
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.
- 3d ago First seen · 141 lines · 12 tokens per session scan A d5803a1c7977
shards is a command published in the GitHub repository geckse/markdown-vdb (23 stars, last pushed 19d ago), licensed MIT. It adds 12 tokens to every session and 1,044 once invoked, about $0.0001 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 commands, from other repositories
check-todos
List outstanding todos and select one to work on.
run-prompt
Delegate one or more prompts to fresh sub-task contexts with parallel or sequential execution.
update
Update Founder OS. Say "update Founder OS" or "pull the latest" (or run /founder-os:update). Pulls the latest System Layer files without touching your personal data, then walks the release packs with you - filtered to the modules you actually run, and proposing every change to a file you have edited as a diff you…
setup
Set up Founder OS from scratch. Fires on any natural-language onboarding ask, including "set up Founder OS", "install Founder OS", "set up my second brain", "help me set up my second brain", "help me onboard", "onboard me", "what do I do", "where do I start", "how does this work", "I'm new", "get me started", "run the…
crux-compress
Compress markdown rule files, code files, and images into CRUX notation for token/size efficiency.
query
Query the OS for a topic. Say "find the file about [topic]" or "search my OS for [topic]" (or run /founder-os:query ). Three modes (index, timeline, full) for progressive disclosure. Read-only.