capture-sweep

capture-sweep is a command for Claude Code from ARCASSystems/FounderOS. It costs 73 tokens per session (358 once invoked), scanned A, original, MIT.

A review step for newly recorded material that sorts each item into client, training, internal, or personal content. It proposes where the items belong but does not file records or route transcripts.

In plain words
What is it for?
Use it to review meeting captures, pasted notes, and recorder content, propose queue items and fact rows, and track what was processed or skipped.
Why use it?
It helps turn scattered meeting and recorder captures into an organised queue while keeping uncertain names and personal material from being treated as confirmed records.

Command for Claude Code

Part of the founder-os plugin — 9 skills, 45 commands, 6 hooks shipped together

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/arcassystems/founderos/capture-sweep
Clone the repo
git clone --depth 1 https://github.com/ARCASSystems/FounderOS

Made for: Claude Code.

Or install founder-os, the plugin that ships this one along with the rest of its 9 skills, 45 commands, 6 hooks.

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.

agentmods badge for capture-sweep

README.md
[![agentmods](https://agentmods.dev/badge/commands/arcassystems/founderos/capture-sweep.svg)](https://agentmods.dev/commands/arcassystems/founderos/capture-sweep)
Your own site
<a href="https://agentmods.dev/commands/arcassystems/founderos/capture-sweep"><img src="https://agentmods.dev/badge/commands/arcassystems/founderos/capture-sweep.svg" alt="Measured on agentmods" height="20"></a>
Per session 73 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 358 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 $0.00073 $0.00358
Opus 5 $0.00036 $0.00179
Sonnet 5 $0.00015 $0.00072
Haiku 4.5 $0.00007 $0.00036

Measured 4d ago against content hash 64bc027303e4, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

capture-sweep 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 4d 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/capture-sweep.md · 26 lines

What it actually says

Founder OS capture sweep

Run the capture-sweep skill at skills/capture-sweep/SKILL.md end to end.

Procedure

  1. Read skills/capture-sweep/SKILL.md.
  2. Preflight the recorder bound in stack.json. If one is named but unreachable, say so in one line, note it in system/quarantine.md, and stop. If none is bound, sweep capture/inbox/ and say plainly that no recorder is connected.
  3. Read the marker at brain/.capture-sweep.json. Missing means first run.
  4. Sort every item into exactly one of the four types and print the decision per item.
  5. Show the proposed queue items and ledger rows, then run them on one go.
  6. Write the marker and close with the honest counts, including what was skipped.

Rules

  • Propose only. Never edit a person's record, never run the capture, never route a transcript.
  • Personal items are skipped and counted, never filed.
  • An unconfirmed name goes to python scripts/unconfirmed_facts.py add, never into an item as a fact.
  • No personal data and no amounts in any item.
  • No em dashes or en dashes. No banned words.
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. 4d ago First seen · 26 lines · 73 tokens per session scan A 64bc027303e4

Subscribe to this mod's changes

capture-sweep is a command published in the GitHub repository ARCASSystems/FounderOS (5 stars, last pushed 20d ago), licensed MIT. It adds 73 tokens to every session and 358 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 2026-08-31.