Add trustworthy_ai_tivm: live trust posture extension - #3
Open
shilawat wants to merge 5 commits into
Open
Conversation
Proposes an optional, additive extension to AgentFacts for continuously-updated, red-team-derived trust scoring, complementing the existing evaluations and certification blocks with a live risk model rather than a static audit snapshot. TIVM methodology from: Trustworthy AI: Red Teaming, Risk and Architecture of Secure Intelligence (ISBN 979-8255167029)
shilawat
force-pushed
the
add-tivm-trust-extension
branch
from
August 13, 2026 06:35
f810da2 to
361a00b
Compare
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Proposes an optional, additive extension to AgentFacts for continuously-updated, red-team-derived trust scoring — complementing the existing
evaluationsandcertificationblocks with a live risk model rather than a static audit snapshot.What this adds
tivm_current: composite risk score (0-100) + SL0-SL5 severity classificationadversarial_bypass_rate: measured red-team attack success ratetrajectory_risk_trend: multi-session behavioral drift detection (rising/stable/declining)trust_tier: T1-T5 assurance tier tied to mission impactexpiry: forces periodic re-attestation rather than a one-time certificationVerification
Validated with ajv against both valid and deliberately-invalid instances — the schema correctly enforces the new constraints.
Example
examples/tivm-extended-agent.jsonis a real AgentFacts document for an agent currently running live on the NANDA network.Compatibility
Fully additive — no existing required fields changed, no breaking changes to any current consumer of this schema.
TIVM methodology proposed by Sandeep Shilawat, author of Trustworthy AI: Red Teaming, Risk and Architecture of Secure Intelligence (ISBN 979-8255167029).