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.
git clone --depth 1 https://github.com/teamtinvio/jaz-aiWrote 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/commands/teamtinvio/jaz-ai/jaz-doc-collect)<a href="https://agentmods.dev/commands/teamtinvio/jaz-ai/jaz-doc-collect"><img src="https://agentmods.dev/badge/commands/teamtinvio/jaz-ai/jaz-doc-collect/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/commands/teamtinvio/jaz-ai/jaz-doc-collect"><img src="https://agentmods.dev/badge/commands/teamtinvio/jaz-ai/jaz-doc-collect.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 | $0.00023 | $0.00596 |
| Opus 5.5 | $0.00009 | $0.00238 |
| Sonnet 5.5 | $0.00005 | $0.00119 |
| Haiku 4.5 | $0.00002 | $0.00060 |
Grade A, and why
jaz-doc-collect 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 1mo 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 — 67 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Document Collection
Run the document collection playbook from the jaz-jobs skill (references/document-collection.md). Scans directories, classifies documents, and prepares them for upload.
Usage
/jaz-doc-collect ~/Downloads/client-docs/
/jaz-doc-collect scan the January invoices folder
Workflow
1. Open the playbook
The steps are in the jaz-jobs skill: references/document-collection.md. Read it first and walk its steps in order. The only clio jobs document-collection command is the ingest sub-tool below.
2. Scan and classify
Point the workflow at a directory:
clio jobs document-collection ingest --source ~/Downloads/client-docs/ --json
The ingestor:
- Scans for PDF, JPG, PNG files
- Classifies each as: invoice, bill, receipt, bank statement, other
- Groups by type and date
3. Upload to Jaz
For invoices and bills, use Jaz Magic (server-side OCR + autofill). The Magic endpoint accepts multipart form data:
- Endpoint:
POST /api/v1/magic/createBusinessTransactionFromAttachment - Fields:
sourceFile(PDF/JPG),businessTransactionType("BILL"or"INVOICE"),sourceType("FILE"or"URL") - Auth: Uses the same sign-in as
clio auth login(or a saved API key); clio handles auth automatically
Magic extraction is async: the upload returns immediately, OCR runs in background. Use clio bills search or clio invoices search to verify the extracted documents.
For bank statements:
clio bank import --account "DBS Current" --file statement.csv --json
4. Verify uploads
Check that uploaded documents appear in Jaz:
clio bills search --from 2025-01-01 --to 2025-01-31 --json
clio invoices search --from 2025-01-01 --to 2025-01-31 --json
Key Rules
- Jaz Magic accepts PDF and JPG/JPEG; use
businessTransactionType: "BILL"or"INVOICE"(not"EXPENSE") - Multipart field names are camelCase:
sourceFile,businessTransactionType,sourceType - Magic extraction is async: the upload returns immediately, OCR runs in background
- Cloud links (Dropbox, Drive, OneDrive) can use
sourceType: "URL"withsourceURLinstead of file upload
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.
- 1mo ago Changed 884895766bff
- 1mo ago Changed · -2 lines 156d4ae5f364
- 1mo ago Changed c23f280b214e
- 2mo ago First seen · 69 lines · 23 tokens per session scan A 6881d3a4afff
jaz-doc-collect is a command published in the GitHub repository teamtinvio/jaz-ai (8 stars, last pushed yesterday), licensed MIT. It adds 23 tokens to every session and 596 once invoked, about $0.0001 per session on Opus 5.5. 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.
jaz-doc-collect compared with commands from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| jaz-doc-collectteamtinvio/jaz-ai | Command | 8 | yesterday | MIT | A | 23 | Claude Code |
| importmichielinksee/bantou | Command | 2 | 2mo ago | MIT | A | 14 | Claude Code |
| reportmichielinksee/bantou | Command | 2 | 2mo ago | MIT | A | 12 | Claude Code |
| create-readmeAzure-Samples/mcp-agent-langchainjs | Command | 204 | 9d ago | MIT | A | 0 | GitHub Copilot |
| journal-entryGAJETOso/financeskills | Command | 47 | 5d ago | MIT | A | 18 | Claude Code |
| reportCPUtester5465/countbean-plugin | Command | 29 | 1mo ago | MIT | A | 26 | Claude Code |
| correctmichielinksee/bantou | Command | 2 | 2mo ago | MIT | A | 13 | Claude Code |