Trust Policy

Last Updated: February 15, 2026

Trust is the foundation of a ratings business. This document describes the structural safeguards, policies, and principles Treebeard uses to maintain independence, prevent conflicts of interest, and ensure the integrity of every rating we publish.

1. Revenue–Rating Separation

Treebeard's revenue operations and rating operations are structurally separated. No amount of money — from subscriptions, API access, Fast Pass fees, or any other commercial relationship — can influence a rating outcome.

1. Rating analysts have no visibility into client revenue or commercial relationships.
2. Revenue teams have no input into rating methodology, scoring, or individual rating decisions.
3. Fast Pass accelerates the review timeline but cannot alter the scoring outcome.

2. Editorial Independence

Rating decisions are made by methodology and analysis teams operating under independent editorial authority. No external party — including investors, partners, rated agents, or marketplace operators — has the ability to influence, preview, or veto a rating before publication.

3. Conflict of Interest Policy

Treebeard team members and contractors are subject to conflict-of-interest restrictions:

1. No team member involved in rating may hold financial positions in rated agents or their tokens.
2. Material conflicts must be disclosed and result in recusal from the relevant rating decision.
3. Former employees of rated organizations are subject to a cooling-off period before participating in that organization's rating.
4. All disclosed conflicts are documented and available for audit.

4. Methodology Transparency

Our methodology is published, our signal categories are documented, and our scoring mechanics are explained in detail on the Methodology pages. We publish this information not because we are required to, but because trust requires visibility into how decisions are made.

1. Signal categories, weights, and scoring mechanics are published on the website.
2. The full methodology is version-numbered and published on the website.
3. Material methodology changes are announced with an explanation of the rationale.
4. Agent type weight profiles and Bayesian calibration parameters are documented.

5. Appeals Process

Any agent team that believes their rating contains an error or is based on inaccurate data may file a formal appeal. Appeals are designed to be fair, structured, and independent:

1. Appeals must be submitted in writing with specific claims and supporting evidence.
2. Appeals are reviewed by an analyst who was not involved in the original rating.
3. Appeal outcomes are documented, including the rationale for upholding or modifying the rating.
4. Appeal decisions are final for the current rating epoch but do not preclude future re-evaluation.
5. Filing an appeal has no negative impact on the agent's current or future ratings.

6. Audit & Accountability

Treebeard maintains internal audit practices and documentation to ensure policy compliance:

1. Every rating decision includes a decision trace documenting the key scoring factors.
2. Methodology changes are version-controlled and timestamped.
3. Conflict-of-interest disclosures are logged and reviewable.
4. We periodically publish transparency reports summarizing rating activity, appeals, and methodology updates.

7. Anti-Gaming Commitment

Treebeard actively works to prevent rating manipulation. Signals are weighted by cost-to-fake, meaning metrics that are easy to game carry less weight. We monitor for coordinated manipulation attempts and reserve the right to flag or downgrade agents engaged in gaming behavior.

8. Contact

Questions about this Trust Policy: trust@treebeardai.com