> **Agent?** Fastest path: MCP at `https://api.mnemom.ai/mcp` — call `get_started` first (zero-auth, no args). Full agent guide: # Acceptable Use Policy - Mnemom [Volver al inicio](/) # Mnemom — Acceptable Use Policy **Last Updated Date: July 7, 2026** Mnemom, Inc. ("**Mnemom**", "**we**", "**us**", or "**our**") provides trust and safety infrastructure for AI agents, including alignment verification, behavioral drift detection, and runtime threat protection (the "**Service**", as defined in our [Terms of Service](/terms)). This Acceptable Use Policy ("**AUP**") describes uses of the Service that are prohibited. This AUP supplements, and is incorporated by reference into, the Terms of Service. Capitalized terms not defined in this AUP have the meanings given to them in the Terms of Service. This AUP applies to all users of the Service, including any AI agents, automated systems, or integrations you operate through or in connection with the Service. ## 1\. Purpose Mnemom's Service is built to make AI agents safer and more accountable. Consistent with that purpose, this AUP prohibits uses of the Service that would turn it toward causing the kinds of harm it exists to prevent. This AUP formalizes prohibitions that Mnemom's runtime protection layer, Mnemom AEGIS, already enforces against inbound and outbound agent traffic (see Section 3 below). ## 2\. Prohibited Uses You shall not use the Service, and shall not permit any agent, integration, or third party acting on your behalf to use the Service, to do any of the following: **2.1 Malicious-Agent Use.** Build, configure, deploy, or operate an AI agent, or use the Service in connection with an AI agent, for the purpose of attacking, compromising, deceiving, or gaining unauthorized access to another party's agents, systems, accounts, or data. This includes, without limitation: prompt injection or indirect injection attacks against other agents; impersonating or spoofing another agent, system, or organization; forging tool outputs, authorization banners, or verification results; social-engineering or business-email-compromise (BEC) schemes directed at automating fraud; and using the Service to probe for, exploit, or evade the security controls of a system you are not authorized to test. **2.2 Election and Democratic-Process Disinformation.** Generate, disseminate, or coordinate the dissemination of false or materially misleading information about the time, place, or manner of voting or elections, voter eligibility or registration, candidates, or election administration, where such information is intended or reasonably likely to deceive voters, suppress lawful voting, or manipulate the outcome of an election. This includes impersonating election officials, election authorities, or candidates, and coordinated inauthentic behavior directed at a democratic process. **2.3 Self-Harm Content.** Generate, promote, provide instructions for, or otherwise facilitate content that encourages or assists suicide, self-injury, disordered eating, or other self-harm. This prohibition does not restrict content that provides support, crisis resources, or recovery information for these conditions. **2.4 Weapons and CBRN Use.** Design, develop, or facilitate the development, acquisition, or use of weapons, or of chemical, biological, radiological, or nuclear ("**CBRN**") materials, agents, or devices, including providing technical uplift, synthesis routes, targeting information, or procurement guidance for any of the foregoing. This prohibition is in addition to, and does not limit, Section 15.11 (Export Control) of the Terms of Service. ## 3\. How Mnemom Enforces This Policy Mnemom AEGIS — the runtime protection layer of the Mnemom Service — screens agent transactions at two checkpoints relevant to this AUP: the **front door** (inbound traffic, before it reaches an agent) and the **back door** (outbound traffic, before it is delivered). The table below maps each prohibited use above to the AEGIS checkpoint(s) and named threat pattern(s) that address it. Prohibited Use AEGIS Checkpoint Threat Pattern(s) 2.1 Malicious-Agent Use Front door Prompt injection; indirect injection; agent spoofing; social engineering; BEC fraud 2.1 Malicious-Agent Use (data theft) Back door Exfiltration patterns; PII leakage 2.2 Election and Democratic-Process Disinformation Front door Social engineering; agent spoofing 2.3 Self-Harm Content Back door Alignment-card violations 2.4 Weapons and CBRN Use Back door Regulated advice; alignment-card violations AEGIS's detection content is continuously updated as new attack patterns are identified; the mapping above reflects Mnemom's enforcement capability as of the Last Updated date of this AUP and does not limit the scope of the prohibitions in Section 2, which apply regardless of whether a particular instance is detected by AEGIS. ## 4\. Reporting Violations If you become aware of a violation of this AUP, please report it to [support@mnemom.ai](mailto:support@mnemom.ai). Security-related reports should be directed to [security@mnemom.ai](mailto:security@mnemom.ai) per Mnemom's [security disclosure process](https://www.mnemom.ai/.well-known/security.txt). ## 5\. Enforcement and Consequences Violation of this AUP is a material breach of the Terms of Service. Mnemom may investigate suspected violations and may, in its sole discretion and without prior notice, remove or disable access to content, suspend or terminate an Account, or take any other action described in Section 12 (Term and Termination) of the Terms of Service. Where Mnemom reasonably believes a violation poses a risk of imminent harm, Mnemom may act immediately and without the notice otherwise contemplated by the Terms of Service. Mnemom may report violations to, and cooperate with, law enforcement or other governmental authorities where appropriate. ## 6\. Relationship to the Terms of Service This AUP is incorporated into, and forms part of, the Terms of Service. In the event of a conflict between this AUP and the Terms of Service, the provision that is more restrictive of the prohibited conduct at issue controls. ## 7\. Changes to This Policy Mnemom may update this AUP from time to time in accordance with Section 15.6 (Agreement Updates) of the Terms of Service. The "Last Updated" date above reflects the date of the most recent revision. --- _Source: /es/acceptable-use-policy/index.html · Generated by build-markdown-mirrors.mjs · For agent-readability commitment #4 see https://www.mnemom.ai/for-agents/_