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 briannaworkman/commonplace --skill designing-commandsgit clone --depth 1 https://github.com/briannaworkman/commonplaceWrote 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/briannaworkman/commonplace/designing-commands)<a href="https://agentmods.dev/skills/briannaworkman/commonplace/designing-commands"><img src="https://agentmods.dev/badge/skills/briannaworkman/commonplace/designing-commands/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/briannaworkman/commonplace/designing-commands"><img src="https://agentmods.dev/badge/skills/briannaworkman/commonplace/designing-commands.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00108 | $0.01437 |
| Opus 5 | $0.00054 | $0.00718 |
| Sonnet 5 | $0.00022 | $0.00287 |
| Haiku 4.5 | $0.00011 | $0.00144 |
Grade A, and why
designing-commands 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 11d 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 — 108 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Designing a command for this vault
Someone wants a command of their own. Your job is to help them design it, not to hand them a standard one. The decisions are the substance of the exercise — a generic command they didn't choose teaches them nothing.
Before you ask anything
Read .claude/commands/ingest.md. That is the house shape: frontmatter with a
description, numbered stages, a check-your-work stage, a report stage. Match it.
How to run this — read this part carefully
Ask every design question in ONE message, as a compact numbered menu with short options.
Do not ask one question at a time. Do not walk them through it conversationally. They are probably in a workshop with minutes to spare, and a five-turn interview will burn the whole block. One message, they reply once, you write the file.
If they answer only some questions, pick sensible defaults for the rest, then say which ones you chose for them so they know what to go change.
If they say "just pick for me" — do it, but afterwards tell them the two choices most worth revisiting and why. They should leave knowing which knobs exist.
Non-negotiables — never put these on the menu
Three things every command in this vault does. They are not choices, so don't offer them as options and don't let a menu answer quietly drop one:
- Read the wiki before answering. Start at
wiki/INDEX.md. - Answer only from what the wiki says. If it isn't there, say so instead of guessing.
- Cite the articles and the
raw/sources beneath them. Article-only citations are not enough — being able to open the original source is the whole point of thesources:field. Whatever output format they pick, the raw paths appear in it.
The questions for /query
Present these as options, with what each one buys. Keep the whole menu under ~15 lines.
- How far should it follow links? Only the articles the index points at / one hop out
along
[[wikilinks]]/ keep going while things look relevant. Shallow is fast and focused; deep finds forgotten connections and sometimes wanders. - Should backlinks count for anything? In Obsidian, what points at an article is as visible as what it points to. An article with many inbound links is probably load-bearing; one with a single inbound link is probably a detail. Options: ignore direction / prefer heavily-linked articles as more central / follow backlinks outward too.
- What happens at the edge of the wiki? Refuse flatly / answer the part it can and mark where the wiki stops / also name the specific source they'd need to ingest. The last one turns every unanswerable question into a reading list.
- Save every answer, or only on request? Always writing to
outputs/builds a record of what they've asked; on-request keeps the folder clean. - How should the citations be presented? Inline links in the prose / a table of articles and their raw sources / a table that also shows the path it walked through the graph. Showing the path is how they catch it reading the wrong articles. Note this is a question about format — that raw sources appear at all is a non-negotiable, not an option.
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.
- 11d ago First seen · 108 lines · 108 tokens per session scan A 4b484e724872
designing-commands is a skill published in the GitHub repository briannaworkman/commonplace (5 stars, last pushed 1mo ago), licensed MIT. It adds 108 tokens to every session and 1,437 once invoked, about $0.0005 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 2026-08-31.
Other skills, from other repositories
remember:ask
Ask a natural-language question of your second brain. The AI reformulates as optimal search queries, calls the mcpremembersearchbrain tool, and synthesizes an answer with citations.
remember:status
Show Remember Second Brain statistics and status.
wiki-ingest
Ingest supplied source material into an Obsidian vault with provenance and claim tracking: pasted text, files staged in the selected vault's inbox or .raw archive, or explicitly approved URLs. Use for a single source or bounded batch, not for saving an assistant answer. Triggers: ingest, ingest this file, ingest this…
autoresearch
Run a bounded, source-grounded research loop, draft a cited dossier, and optionally propose a separately reviewed canonical vault merge. Use when the user wants autonomous or deep research that may access the public web. Triggers: /autoresearch, autoresearch, research this topic, deep dive into, investigate, find…
canvas
Create, inspect, and update Obsidian JSON Canvas boards with text, file, link, group, and edge nodes. Use for canvas status, canvas lists, visual maps, zones, spatial layouts, adding vault notes or media to a .canvas file, and requests such as create canvas, add to canvas, or put this on the canvas.
save
Save a user-selected answer, decision, insight, or session summary into an Obsidian vault as one reviewed transaction. Use only when the user explicitly asks to preserve specific conversation content, not when they supply a file or URL to ingest. Triggers: /save, save this, save that answer, file this conversation…