Usage
Policy

Purpose and scope

This Usage Policy defines the boundaries for using Oscerra responsibly. It applies to every Oscerra service, agent, execution surface, preview feature, account, integration, and tool we make available, whether access is free, paid, invited, experimental, or provided under a commercial agreement.

Autonomous systems can convert an instruction into real-world action. That makes misuse qualitatively different from a bad answer in a chat window. The policy therefore focuses not only on what a user asks for, but also on what the system is being authorized to access, change, publish, purchase, communicate, or execute.

Core rule. Do not use Oscerra to obtain, automate, scale, conceal, or operationalize conduct that you would not be lawfully and ethically permitted to carry out yourself.

Authorization and access boundaries

Only direct Oscerra to access systems, accounts, data, devices, websites, repositories, workspaces, or credentials that you are authorized to use for the requested purpose. Authorization must cover the action, not merely the ability to reach the target.

• Do not use stolen, leaked, shared-without-permission, or fraudulently obtained credentials.
• Do not bypass authentication, access controls, paywalls, account restrictions, technical safeguards, or anti-abuse mechanisms for an unauthorized purpose.
• Do not use automation to evade a platform ban, rate limit, enforcement action, or contractual restriction through account rotation, identity manipulation, or similar techniques.
• Allowed: good-faith operation of systems you own or are authorized to administer, including ordinary login flows, testing environments, and approved automation within the permissions you have been given.

Cybersecurity and technical abuse

Oscerra may be used for defensive security, authorized testing, software engineering, incident response, and research where you have permission to perform the work. It may not be used to compromise systems, steal secrets, persist in someone else’s environment, conceal malicious access, or materially facilitate cyber harm.

• No malware deployment, ransomware, destructive payloads, credential theft, session hijacking, botnet operation, or unauthorized persistence.
• No phishing infrastructure, deceptive credential collection, malicious impersonation, or automated social engineering intended to obtain secrets or unauthorized access.
• No unauthorized scanning or exploitation of third-party systems where the activity creates meaningful intrusion, disruption, exfiltration, or evasion risk.
• No instructions whose primary purpose is to defeat security monitoring, forensic attribution, abuse controls, or law-enforcement detection for malicious activity.
• Authorized vulnerability research, capture-the-flag environments, sandboxed analysis, and remediation work are permitted when they are genuinely authorized and proportionate.

Privacy, surveillance, and personal data

Do not use Oscerra to unlawfully monitor people, aggregate sensitive personal information, identify anonymous individuals, expose private data, or make intrusive inferences that a reasonable person would not expect. Having technical access to information does not necessarily give you the right to collect or repurpose it.

• No doxxing, stalking, non-consensual tracking, or publication of private contact, location, financial, medical, identity, or family information for harassment or harm.
• No unlawful interception of communications, covert access to private accounts, or spyware-like monitoring.
• No face recognition, biometric identification, or sensitive-trait inference where prohibited by law or where used to enable discrimination, surveillance, or coercion.
• Legitimate administrative, security, fraud-prevention, accessibility, and user-requested data workflows are allowed when they are lawful, proportionate, and appropriately secured.

Fraud, impersonation, deception, and manipulation

Do not use Oscerra to deceive people about identity, authority, provenance, or material facts in order to obtain money, credentials, access, votes, confidential information, legal advantage, or other benefits. Automation must not be used to make a fraudulent scheme cheaper, faster, or harder to detect.

• No financial fraud, fake invoices, account-takeover schemes, romance scams, investment scams, refund fraud, or identity theft.
• No impersonating a real person or organization in a materially deceptive way, including synthetic communications designed to induce reliance on false authority.
• No fabrication of evidence, records, receipts, citations, identities, testimonials, or credentials for use as if they were genuine.
• Parody, fiction, simulation, testing, and clearly disclosed synthetic content are generally allowed when they are not used to mislead people about a material real-world fact.

Spam, platform manipulation, and abusive scale

Do not use Oscerra to mass-message, mass-post, mass-register, mass-review, mass-vote, or otherwise manipulate online systems in ways that violate law, consent, or platform rules. Volume can transform individually harmless actions into abuse.

• No unsolicited bulk messaging, robocall-style abuse, fake engagement, review manipulation, astroturfing, or coordinated inauthentic behavior.
• No automated account creation or content submission intended to overwhelm moderation, ranking, marketplace, ticketing, reservation, or allocation systems.
• Legitimate customer communications, internal operations, support workflows, and consent-based outreach are allowed when they comply with applicable anti-spam and privacy law.

High-impact decisions and human responsibility

Do not use Oscerra as the sole or final decision-maker for determinations that can materially affect a person’s legal rights, safety, liberty, health, employment, housing, credit, insurance, education admission, access to essential services, or comparable high-impact interests unless the use is lawful and includes meaningful, appropriately qualified human review.

The Service may assist with research, organization, drafting, triage, or workflow support in these domains. The prohibition is on removing the accountable human and treating probabilistic output as sufficient authority for a consequential decision.

Violence, weapons, and physical harm

Do not use Oscerra to plan, facilitate, optimize, conceal, or execute violent wrongdoing or to procure, construct, or deploy weapons for harmful or illegal purposes. Do not use autonomous execution to control physical systems in a way that creates unreasonable risk of injury or property damage.

Benign safety education, historical analysis, lawful sport, high-level policy discussion, and protective or emergency-preparedness information may be allowed. The relevant question is whether the assistance meaningfully increases a person’s ability to cause harm.

Sexual exploitation and protection of minors

Oscerra must never be used to create, obtain, transform, distribute, advertise, facilitate, or sexualize child sexual abuse material or other sexual exploitation involving minors. Do not use the Service to groom, traffic, blackmail, extort, or otherwise exploit any person sexually.

We may take immediate protective action, preserve relevant records where legally required, and report apparent child sexual exploitation or other legally reportable abuse to appropriate authorities or organizations.

Self-harm and dangerous encouragement

Do not use Oscerra to encourage, coach, optimize, or operationalize self-harm, suicide, dangerous starvation, or other conduct intended to cause serious injury to a person. Supportive, recovery-oriented, educational, and help-seeking use is permitted.

Where a workflow suggests an immediate risk to life or safety, do not rely on autonomous software as the sole response. Contact appropriate emergency services, crisis resources, trusted people, or qualified professionals.

Regulated goods, financial activity, and other controlled conduct

Do not use Oscerra to unlawfully purchase, sell, distribute, broker, advertise, or evade controls for regulated or prohibited goods and services. This includes illegal drugs, unlawfully traded weapons, exploitative sexual services, stolen property, fraudulent financial instruments, sanctions evasion, and other transactions prohibited by law.

Lawful commerce in regulated industries may require identity verification, licensing, age controls, transaction monitoring, professional review, or other safeguards. Users are responsible for implementing those obligations rather than treating automation as a way around them.

Information integrity, research, and public claims

Do not use Oscerra to manufacture false evidence or coordinate deceptive influence operations. If you use the Service for research, reporting, analysis, or public communication, verify consequential facts and preserve meaningful uncertainty rather than presenting generated material as verified merely because it is fluent.

Synthetic content should be disclosed when the context creates a meaningful risk that people will mistake it for an authentic record, statement, endorsement, or event. We may introduce provenance or disclosure features for particular outputs and expect users not to remove them for deceptive purposes.

Protecting Oscerra and other users

Do not interfere with the Service, probe it in a way that creates unreasonable operational risk, extract secrets, bypass billing or rate limits, automate account abuse, resell access without permission, or attempt to obtain internal model prompts, credentials, private user data, or non-public system information through exploitation.

Good-faith security research should be conducted through a scope and process authorized by Oscerra. If you discover a vulnerability, report it to support@oscerra.space and avoid accessing data or systems beyond what is necessary to demonstrate the issue safely.

Autonomy-specific safeguards

When a task can take external action, users must exercise additional care. The following patterns are not acceptable unless specifically authorized, lawful, and appropriately controlled:

• Giving an agent broad administrative credentials when narrow permissions would be sufficient.
• Running irreversible or high-impact workflows without a review or rollback plan where one is reasonably available.
• Using repeated automated attempts to pressure, overwhelm, or circumvent another system’s safeguards.
• Instructing the Service to conceal its activity from a system owner, user, auditor, or person entitled to know about the action.
• Treating a successful tool call, browser state, or execution receipt as proof that the underlying decision was ethically, legally, or factually correct.

Enforcement and proportionality

We may refuse an individual request, limit a capability, require extra confirmation, quarantine a workflow, suspend an account, or terminate access when we reasonably believe use violates this policy or creates material risk. We may also act to comply with law or valid legal process.

Enforcement should be proportionate. A failed experiment, accidental policy edge case, or misunderstood instruction is not the same as deliberate abuse. Where appropriate, we may provide a warning, explain the category of concern, and allow a user to change the workflow. Immediate action may be necessary for severe abuse, security threats, exploitation, or imminent harm.

Reporting abuse and policy questions

If you believe Oscerra is being used in a way that violates this policy, or you need to understand whether a legitimate workflow is permitted, contact support@oscerra.space. Include enough context for us to understand the activity, but do not send passwords, authentication codes, private keys, or unnecessary personal information.

Updates to this policy

We expect this policy to evolve as autonomous systems gain new capabilities and as real-world failure modes become better understood. We may update it to clarify ambiguous cases, address new abuse patterns, or add safeguards. Material changes will be published on this page with an updated effective date.

A policy that never changes while the technology changes would not be a serious policy. Our goal is to keep the boundaries understandable, enforceable, and connected to actual product behavior rather than accumulating vague prohibitions that users cannot operationalize.