This document describes how QUILL handles privacy for local and AI-assisted workflows.
When you use cloud AI providers, the prompt content you choose to send is transmitted to that provider. Provider-side storage, retention, and policy behavior are controlled by that provider's terms and settings, not QUILL.
QUILL does not persist Ask Quill chat transcripts or Writing Assistant interaction transcripts by default. If you explicitly copy output into a document, that content is then part of your document and saved according to your normal file and backup workflow.
QUILL stores API keys using Windows Credential Manager when available. If Credential Manager is unavailable, QUILL falls back to DPAPI-encrypted local secret storage.
QUILL does not store API keys in plaintext.
QUILL may create local settings and state files under your app data
directory (for example %APPDATA%\Quill\...), including:
These files are local to your machine and are not uploaded by default.
If you use the QUILL Developer Console (QDC), each command you run is
appended to a history.jsonl file in the app-data directory
(up to 500 entries; oldest are removed first). Entries are passed
through QUILL's redaction layer before being written, so API keys and
tokens that match known patterns (GitHub PATs, OpenAI keys, AWS access
keys, Slack tokens, and long alphanumeric tokens) are replaced with
[TOKEN] in the history file.
If you believe sensitive data was stored before the redaction layer
was active, you can delete the history.jsonl file from the
app-data directory manually.
When you open a file from a GitHub repository using File >
Open from Remote, QUILL downloads the file into a
github-temp subdirectory of the app-data folder. These
files are not automatically deleted when you close the tab or exit
QUILL.
If you work with private repositories, review the
github-temp directory periodically and delete files you no
longer need. The directory is local to your machine and is not shared or
uploaded.
You are responsible for reviewing AI-generated output before using, sharing, or publishing it. For sensitive content, use local models when possible and verify that cloud use meets your organization's security and compliance requirements.