How it works

From approved evidence to a reviewed response.

You supply the questionnaire and approved answers. The workbench prepares a response, records its sources and separates the questions that need your team.

01 / Prepare

Make an answer library your team can maintain.

Start with the blank answer-library template, or import a previously completed Excel questionnaire. Library Builder extracts existing answers into an approval queue. Importing an answer does not approve it.

Check each statement against your evidence. Record its owner, applicability, source reference and approval expiry. The library becomes the source for matching, drafting and final checks.

Library setup instructions →

02 / Prepare a response

Work with the file your buyer sent.

Select the question and response columns in a CSV or Excel questionnaire. For Excel, the app writes a new copy of the workbook, retains its sheets and layout, and leaves existing human responses in place.

A matching approved answer is copied verbatim. When drafting is enabled, related approved answers can support a proposed draft. The claim guard checks that draft against its cited answers. Unsupported drafts are discarded.

Matched
An approved answer was copied.
Drafted or suggested
A draft or related library entries need your review.
Needs evidence
The library does not cover the question.
Legal review
A person needs to review a contractual or legal question.
Flagged or blocked
Suspicious input or an unsupported draft needs attention.
Workbook instructions →

03 / Review

Check the file you will actually send.

Review responses alongside their cited evidence. Edits are checked as you type. Export the final file, then use Check before sending to compare the final answers with the current approved library.

The checks look for unsupported certification terms, changed numbers and frequencies, lost negation, broader scope and expired approvals. You can add your own guard terms for claims specific to your business.

The tool checks wording against approved answers. It does not establish whether an approval is accurate, whether a control is operating, or whether a response satisfies the buyer’s requirements. A person must resolve findings and approve the file.

Pre-send instructions →

Files you keep

Responses, review notes and sources.

  • Completed workbook: a separate copy of the buyer’s Excel file with proposed answers and review statuses.
  • Response draft: CSV or XLSX containing questions, answers, statuses and evidence references.
  • Review queue and missing evidence: CSV lists for your reviewers and evidence owners.
  • Review report: an HTML document grouping questions by the attention they need.
  • JSON results: structured output for your own tooling.
  • Pre-send report: HTML and CSV findings, plus possible additions to your library.

Trust Pack Builder separately produces a customer assurance HTML page and a JSON manifest from answers explicitly marked shareable. You review and distribute those files yourself.

04 / Security program

Run the rest of your security work in the same app.

  1. Describe your company. The data you hold, where you operate and which rules apply decide the frameworks, notification duties and policy wording.
  2. Assess 52 safeguards once. Readiness for 16 frameworks, from SOC 2, ISO 27001 and NIST CSF 2.0 to GDPR, CMMC and ISO 42001, is calculated requirement by requirement. Your approved answers can mark the safeguards they describe as partly in place.
  3. Collect evidence. Attach files to each safeguard and let the weekly checks file their own results: domains, GitHub, MFA and devices.
  4. Write and approve policies. Seventeen policies are written from your profile and safeguards, as Word documents with framework references. Record which staff signed each version and who completed training.
  5. Track findings and access. Import scanner exports for fix-by dates, and review who has access to each system from its user list.
  6. Track risks and vendors. Risks are suggested from your gaps; vendors are tiered by data and access and sent a questionnaire sized to the tier.
  7. Respond to incidents. A playbook for the kind of incident, a timeline, and every notification deadline counting down.
  8. Report. A posture report for leadership, a readiness workbook with the ISO 27001 Statement of Applicability, and one audit package with every document and evidence file.

Readiness reflects your own assessment. It prepares you for an audit or assessment; it does not certify your company.

Security program instructions

Data & optional AI

Choose where the model runs.

Without AI

Matching, evidence gap lists, legal routing and checking run on your computer without a model or provider account.

With a local model

Connect llama.cpp, Ollama or LM Studio on the same computer. You install and run the model separately.

With a cloud provider

Connect your own provider key. Cloud drafting sends the question and up to three matching approved answers to the selected provider after confirmation. Provider charges and terms apply.

Intellinaire chat

Chat uses your chosen model and app context. Attachments and web search can disclose additional content to the services you enable. Proposed changes require approval in the app.

Read the privacy notice and the cloud setup instructions before enabling a service.

Next

Check the requirements.

Pricing & downloads →