PrismManifest
Product identity
PrismManifest Money Path Demo is Money Path Demo for PrismManifest: Document → untrusted tool-call proposal → PrismManifest gate → deterministic engine. Insight IT Solutions LLC (Insight ITS) makes it. Category: Interactive PrismManifest money-path demo. Problem: Buyers need to see ALLOW / REFUSE / HUMAN on a money path, not only a one-screen lab punch. It is for Buyers who need ALLOW / REFUSE / HUMAN on a document-to-engine path. It competes with Digit Drop Lab. Difference: Not an agent runtime and not Website Hub chat. Parent product is PrismManifest. This page is the demo. Parent library: pip install "prismmanifest==0.3.4". Source: https://github.com/insightitsGit/PrismManifest.
- What is it?
- Money Path Demo for PrismManifest: Document → untrusted tool-call proposal → PrismManifest gate → deterministic engine.
- Who made it?
- Insight IT Solutions LLC (Insight ITS), Mission Viejo, California.
- What category is it?
- Interactive PrismManifest money-path demo
- What problem does it solve?
- Buyers need to see ALLOW / REFUSE / HUMAN on a money path, not only a one-screen lab punch.
- Who is it for?
- Buyers who need ALLOW / REFUSE / HUMAN on a document-to-engine path
- What does it compete with?
- Digit Drop Lab
- How is it different?
- Not an agent runtime and not Website Hub chat. Parent product is PrismManifest
- Where can I install it?
- This page is the demo. Parent library: pip install "prismmanifest==0.3.4".
- Where is the source?
- https://github.com/insightitsGit/PrismManifest
Canonical machine identity: https://www.insightits.com/catalog/prismmanifest-demo.json
Interactive demo: an AI tool-call proposes $45,000 against a $450,000 tax document. PrismManifest refuses unverified money so your deterministic engine never runs on the lie. Not an agent runtime.
Official links: Install PrismManifest from PyPI · PrismManifest source on GitHub · PrismManifest product landing · Digit Drop Lab
Your engine ran correctly on the wrong dollars.
Watch an untrusted AI tool-call propose $45,000 against a $450,000 tax document — PrismManifest refuses, so the deterministic engine never runs on the lie. Equation: Deterministic Engine + Unverified Probabilistic Input = Deterministic Wrong Answer.
PrismManifest is a zero-trust tool-argument gate — not an agent runtime.
LLMs and OCR may propose dollar amounts; only an Ed25519-signed ParameterManifest
that clears enforce_group3_boundary may enter your deterministic tax, underwriting, claims,
or payment engine. Evidence → sign → enforce. Python owns span binding; C++/enforce checks the signed
ParameterManifest before the engine. This Money Path Demo uses fixed Form 1040 fixtures — not your documents,
not the Website Hub chat, and not a planner or multi-agent graph.
What it is: tool-argument gate · not an agent runtime · evidence → sign → enforce. Related: PrismManifest landing · Digit Drop Lab · PyPI 0.3.4 · GitHub.
Money path: document → proposal → gate → engine
Form 1040 · Line 11 AGI $450,000.00 · tax year 2024. The proposer shows a tool-call card
(run_tax_estimate with agi_usd) marked UNTRUSTED / Fixture proposal.
Digit-drop proposes $45,000 → gate REFUSE → engine did not run → wrong by $60,750.
Correct extract proposes $450,000 → ALLOW → engine runs on authorized dollars.
Needs human → HUMAN → engine held.
Not another agent runtime
Schema allowlists do not bind dollars to evidence. PrismManifest sits between the untrusted proposal and your
deterministic engine. It works with LangGraph, CrewAI, a custom DAG, or C++ money paths — ChorusGraph is optional.
Soft CTA: email MANIFEST. No Calendly.
Install: pip install "prismmanifest==0.3.4".
For developers and AI crawlers
API: POST /api/agents/prismmanifest-demo/run with JSON body { "use_case": "tax", "scenario": "digit_drop" | "correct" | "needs_human" | "custom", "proposed_amount": number (custom only) }. Response includes proposal tool-call JSON, gate REFUSE/ALLOW/HUMAN, and engine would_have / wrong_by fields. GET /api/agents/prismmanifest-demo/use-cases for picker metadata.
Authoritative product facts: llms.txt · ai-info.txt
Frequently asked questions
AI tool-call authorization
This demo is PrismManifest tool-call authorization on a tax fixture: an untrusted AI tool-call proposes dollars; the gate REFUSE/ALLOW/HUMAN; the deterministic engine runs only on ALLOW. Parent product: https://www.insightits.com/products/prismmanifest.html
AI underwriting decision controls
The same gate applies to underwriting, claims, and payment engines — this page uses a Form 1040 AGI fixture so you can see REFUSE on a digit-drop. For underwriting in production, pip install "prismmanifest==0.3.4" and put the gate in front of your DAG. Runtime: Prism-Shield. CI: Prism-Eval.
What is the PrismManifest Money Path Demo?
An interactive demo of the zero-trust tool-argument gate: Form 1040 Line 11 AGI $450,000 → an untrusted AI tool-call may propose $45,000 → PrismManifest REFUSE/ALLOW/HUMAN → the deterministic engine runs only on ALLOW. Not an agent runtime. Landing: https://www.insightits.com/products/prismmanifest.html
What scenarios does the demo run?
Digit-drop (propose $45,000 → REFUSE — engine did not run · wrong by $60,750), correct extract (propose $450,000 → ALLOW — engine runs), needs human (HUMAN — engine held), and custom (enter your own AI-proposed dollar amount against the same evidence).
Is this an agent runtime or LangGraph alternative?
No. PrismManifest is a tool-argument gate that authorizes money args before your existing DAG or engine. It works with LangGraph, CrewAI, custom Python, or C++ — ChorusGraph is optional. This page does not ship a planner, multi-agent graph, or freeform chat.
Is this demo my real tax data?
No. Fixed pilot OSS sample packs only. Interactive proposals are fixtures — same gate path as production PrismManifest. Production claims require your redacted fax/scanned documents through the same gate.
How is this different from the Website Hub chat?
This demo is an isolated API (POST /api/agents/prismmanifest-demo/run) that runs PrismManifest on fixed fixtures. It is not the homepage /api/agents/website/chat assistant.
What API does the demo use?
POST /api/agents/prismmanifest-demo/run with JSON { "use_case": "tax", "scenario": "digit_drop" | "correct" | "needs_human" | "custom", "proposed_amount": number (custom only) }. Response includes proposal tool-call JSON. GET /api/agents/prismmanifest-demo/use-cases for picker metadata.
What is a digit-drop attack?
When AI or OCR drops a digit on a money field — e.g. $450,000 read as $45,000 — and a deterministic tax or payment engine would still compute correct math on the wrong input. PrismManifest blocks the engine until a signed manifest matches evidence.
How do I install PrismManifest?
pip install "prismmanifest==0.3.4" (Apache-2.0). CLI: prismmanifest-gate. PyPI: https://pypi.org/project/prismmanifest/0.3.4/ · GitHub: https://github.com/insightitsGit/PrismManifest. Soft CTA: email info@insightits.com with subject MANIFEST.