A command that turns new files in a raw source folder into organised wiki articles. It checks which Markdown files have not been processed, reads them according to their type, and follows the vault's article and metadata rules.

Importing notes, fetched web pages, and cleaned transcript material from raw/ into wiki/ articles with the required metadata.

It removes the repetitive work of finding new sources and turning notes, links, and transcripts into consistent articles. It also filters noise from transcripts and avoids processing the same source twice.

ingest — Command for Claude Code by briannaworkman

Install

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.

Clone the repo

git clone --depth 1 https://github.com/briannaworkman/commonplace

Made for: Claude Code.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/commands/briannaworkman/commonplace/ingest.svg)](https://agentmods.dev/commands/briannaworkman/commonplace/ingest)

Your own site

<a href="https://agentmods.dev/commands/briannaworkman/commonplace/ingest"><img src="https://agentmods.dev/badge/commands/briannaworkman/commonplace/ingest.svg" alt="Measured on agentmods" height="20"></a>

Per session 9 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 683 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

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	$0.00009	$0.00683
Opus 5	$0.00005	$0.00342
Sonnet 5	$0.00002	$0.00137
Haiku 4.5	$0.00001	$0.00068

Measured 8d ago against content hash 94ad9166f91f, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

ingest 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.

finished/.claude/commands/ingest.md · 65 lines

How it starts

The opening of the file, as written. The whole thing — 65 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Compile every un-ingested source in raw/ into wiki/ articles.

CLAUDE.md defines the frontmatter schemas, the naming rule, and what an article is. Follow those. Everything below is the procedure.

1. Find what's new

List every .md file under raw/. For each path, search wiki/ for that path appearing in a sources: list. Any path with no match is un-ingested.

Report the list before doing any work. If nothing is un-ingested, say "Nothing to ingest" and stop.

2. Read each source, according to its type

note — read as-is.
link — if the file contains only a bare URL, fetch the page and replace the body with the fetched text, keeping the frontmatter. If the fetch fails, leave the file alone, say so, and move on — the next run will retry it.
transcript — read it, but strip the noise on the way out: speaker labels, timestamps, filler, crosstalk, tangents, and ad reads never appear in an article. Leave the file in raw/ exactly as it is. The mess is the archive; the article is the understanding. Most of a transcript is not an idea — expect a long one to yield two or three articles, and do not pad to fill space.

Add frontmatter to any source missing it. Do not otherwise touch a source's body.

3. Map the ideas onto the existing wiki

Read wiki/INDEX.md. For each idea in the source, decide: does an article already cover this? Prefer updating an existing article over creating a near-duplicate.

4. Write the articles

One article per concept. Add [[wikilinks]] in both directions — link the new article to related ones, then edit those to link back.

5. Record provenance

Append the source's path to the sources: list of every article you touched. This is what makes the source count as ingested, so it is not optional.

6. Update the index

Add one line per new article to wiki/INDEX.md.

7. Check your work before reporting

Every article's filename is its exact title, Title Case, with spaces. Do not slugify.
Every [[link]] you wrote points at a title that has a file.
No [[link]] has a line break inside it. Prose wrapping does this silently, a split target does not resolve, and if the article also lists that link in related: the graph still looks fine — so nothing will alert you. Search your own output for [[ followed by a newline before the closing ]], and re-wrap those lines.
Every article you touched lists the source in sources:.

Read the full file on GitHub · 65 lines

Changes

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 · 65 lines · 9 tokens per session scan A 94ad9166f91f

Subscribe to this mod's changes

ingest is a command published in the GitHub repository briannaworkman/commonplace (5 stars, last pushed 27d ago), licensed MIT. It adds 9 tokens to every session and 683 once invoked, about $0.0000 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.

Claude Code Documents knowledge-management markdown note-taking obsidian second-brain

ingest