Add prompt injection and guardrails security guide#1113
Draft
veltson-vapi wants to merge 1 commit into
Draft
Conversation
veltson-vapi
force-pushed
the
docs/prompt-injection-and-guardrails
branch
from
July 21, 2026 23:15
404a00e to
42e3327
Compare
Contributor
|
🌿 Preview your docs: https://vapi-preview-019f86ff-08d5-757a-bc1c-edb73b9b00f6.docs.buildwithfern.com |
veltson-vapi
force-pushed
the
docs/prompt-injection-and-guardrails
branch
from
July 21, 2026 23:35
42e3327 to
9efd059
Compare
Contributor
|
🌿 Preview your docs: https://vapi-preview-019f8712-e5ad-707d-8ae5-8fb12ea6c057.docs.buildwithfern.com |
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
veltson-vapi
force-pushed
the
docs/prompt-injection-and-guardrails
branch
from
July 21, 2026 23:50
9efd059 to
db030ff
Compare
Contributor
|
🌿 Preview your docs: https://vapi-preview-019f8725-83ae-743a-886e-51496c0cc1b0.docs.buildwithfern.com |
Contributor
|
🌿 Preview your docs: https://vapi-preview-019f8731-e3c5-72ab-8980-2039a4c86a15.docs.buildwithfern.com |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Description
Adds a new Prompt injection and guardrails guide under Security and privacy. It covers the layers available to protect an assistant against prompt injection and data exposure.
fern/security-and-privacy/prompt-injection-and-guardrails.mdxfern/docs.ymlcompliancePlan.securityFilterPlan), then keeping sensitive data out of context (the return-a-status pattern, stored credentials, and tool argument encryption)securityFilterUtil.ts,security.types.ts) and the existing tool-argument-encryption guideOpen questions for review
sanitizevsreplacefilter modes. InsecurityFilterUtil.tsthese run identical code (both substitute the matched phrase withreplacementText). The doc currently documentssanitizeand notes thatreplacebehaves identically. Is that the intended behavior, or shouldsanitizestrip to empty whilereplacesubstitutes? The original author (Adi) has left, so this needs a second opinion before we commit to the wording.prompt-injectionregexes. That is useful for honest users, but it also tells an attacker exactly how to word around the filter. Keep them, or describe the categories and link the API reference instead?Testing Steps
fern docs devor navigate to preview deployment