process-inbox

A command for processing files placed in a drop folder and turning them into organized knowledge files. It archives originals, assigns topics, and records where each file came from.

In plain words
What is it for?
Useful for importing notes, PDFs, images, data files, and unstructured text into a structured knowledge repository, including cross-referencing related material.
Why use it?
It removes the manual work of sorting incoming files while preserving the original files and their sources.

Command for Claude Code

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.

agentmods
npx agentmods add commands/lanzelot1/claw-brain/process-inbox
Clone the repo
git clone --depth 1 https://github.com/Lanzelot1/claw-brain

Made for: Claude Code.

Per session 8 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 666 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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.

ModelPer sessionOnce invoked
Fable 5 $0.00008 $0.00666
Opus 5 $0.00004 $0.00333
Sonnet 5 $0.00002 $0.00133
Haiku 4.5 $0.00001 $0.00067

Measured 3d ago against content hash fc89ac14309f, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

process-inbox 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.

.claude/commands/process-inbox.md · 42 lines

How it starts

The opening of the file, as written. The whole thing — 42 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Process all new files in the drop/ folder.

Steps

  1. Read memory/_index.md to discover all available areas
  2. For each file in drop/: a. Read the content and understand what it's about b. Archive original: Copy the file to raw/ (prepend YYYY-MM-DD- if no date prefix exists). Never overwrite — if a file with that name exists in raw/, append a counter. c. Categorize: Match it to an existing area from the index d. Prepare knowledge file:
    • If the file is already a well-structured .md with frontmatter: move it to knowledge/{area}/, ensure source: frontmatter exists, add type: source if no type: field.
    • If the file is a PDF, image, data file, or unstructured text: create a new .md file in knowledge/{area}/ that extracts the key facts, data, and quotes. Set source: to the raw/ path. Set type: source. The raw file stays in raw/ only. e. Check source: frontmatter: If the source is unclear, ask the user: "Where did this file come from?" f. Cross-reference: Search other knowledge areas for substantively related files. If found, ask: "I found these related files — should I add cross-references?" If yes, add ## Related sections with reciprocal links. g. Update wiki pages: Check if any existing wiki pages (type: wiki) in knowledge/ cover topics related to this new source. If so, ask: "These wiki pages may need updating with the new information: [list]. Update them?" If yes, incorporate new facts and add the new source to their ## Sources section. h. Add entries to memory/_index.md (area section + Recently Added) i. Log: Append to memory/log.md: YYYY-MM-DD HH:MM | ingest | {filename} → knowledge/{area}/{filename}
  3. If a file doesn't fit any existing area:
    • Ask the user: "This file doesn't fit any existing area. Want to create a new one?"
    • If yes, run the /new workflow first, then process the file into the new area
  4. Remove processed files from drop/

Read the full file on GitHub · 42 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.

  1. 3d ago First seen · 42 lines · 8 tokens per session scan A fc89ac14309f

Subscribe to this mod's changes

process-inbox is a command published in the GitHub repository Lanzelot1/claw-brain (11 stars, last pushed 4mo ago), licensed MIT. It adds 8 tokens to every session and 666 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 2026-08-30.