Submit
Upload or paste the text
Submit a .doc, .docx, .txt, or .pdf file, or paste text directly. 300 or more words of English gives the most reliable result.
Dechefr is a structured professional judgment tool that assesses written or transcribed communication for the risk of targeted violence. It evaluates emails, posts, letters and messages against twelve research-grounded indicators, producing a result ready for a trained assessor to review.
Need to manage the whole case, not just one text? Dechefr also runs inside Sentinel, our guided solution for reporting, assessing, and managing conflict, threats, and violence.
What the words reveal
Written communication can carry indicators that bear on risk, from grievance and fixation to leakage and last-resort escalation, and reading them consistently takes a defined method rather than an experienced eye alone. Dechefr, from the French déchiffrer, to decipher, is a structured professional judgment (SPJ) tool built for that. It assesses a piece of writing or a transcript against twelve research-based indicators drawn from the literature on warning behaviors and the pathway to violence, and returns each indicator with an overall risk level. As with any SPJ instrument, the result structures the assessment rather than settling it. The judgment stays with a trained assessor.
Dechefr — assessment
From a piece of writing to an evidence-based report
Dechefr turns a document or pasted text into a structured, transparent assessment in four steps.
02Assess
Review the risk level and indicators
A weighted overall risk level gives a fast triage signal across four levels, from no indicators present to several significant indicators of potentially violent behavior. Review each of the twelve indicators with an interactive assistant, and adjust any call based on your own judgment.
03Compare
Run the similarity assessment
See how the case compares against a reference database of violent offenders' communications and against your own case archive, showing which patterns recur and which details are particular to this case.
04Report
Generate an evidence-based report
A structured report documents every indicator, the sentences behind it, and the overall risk level, giving you a record of how the assessment was reached.
What Dechefr looks for
Dechefr assesses the text and only the text. It is not swayed by who wrote it, by internal pressure, or by the fatigue that builds over a long caseload, and it applies the same criteria to the hundredth message as to the first. That makes it a stable second opinion for assessors to weigh their own judgment against, whether it runs on its own or inside Sentinel.
Dechefr looks for twelve indicators, drawn from the literature on warning behaviors and the pathway to violence.
01 · Anger
Expressions in the communication that indicate anger.
02 · Grievance
An unmet need that fuels a sense of being wronged, the kind that can drive revenge or retribution.
03 · Othering
Us-versus-them thinking that marginalizes a "them" group as an outgroup.
04 · Leakage
Communication of intent to harm a specific target.
05 · Influences from previous offenders
Interest in or admiration for previous offenders or attacks.
06 · Warrior mentality
Military jargon and the adoption of a warrior perspective.
07 · Preoccupation
Extreme occupation with a person or cause.
08 · Linguistic alignment
Linguistic markers of influence from an ideology or subculture.
09 · Preparation activities
Violent ideation being converted into operational capability.
10 · Last resort and desperation
Violence perceived as the only remaining option, with urgency and closing alternatives.
11 · End-of-life and finality
Anticipation or acceptance of death, capture, or irreversible consequences.
12 · Legacy and martyrdom
Violence framed as historic, justified, or worthy of remembrance.
Use Dechefr on its own, or the full platform
The same assessment engine and the same twelve indicators, whether Dechefr stands alone or sits inside Sentinel. Start with the scope your team needs today, and add the case-management layer when you're ready.
Assess the communication
For analysts, forensic linguists, and researchers who need to assess a single piece of written communication.
- Paste or import text from any source: social media, forums, email, letters.
- Get an overall risk level and the twelve research-based indicators, with the passages behind them.
- Keep results in your personal case archive for comparison and tracking.
- Runs in the browser. Sign in and start assessing, with no case-management setup.
Manage the whole case
For teams that receive reports and manage cases end to end, not just single texts.
- Structured intake, one queue for every concern, ready for triage.
- Structured assessments, role-based access, full audit trail, and monitoring over time.
- Dechefr assists automatically whenever a case contains communication, writing its result into the case record.
- One place from first report to closure, so nothing falls between systems or shifts.
Built by researchers, used by practitioners
Mind Intelligence Lab was founded by researchers in threat assessment and computational text analysis. Our tools are grounded in the research literature on targeted violence, structured professional judgment, and warning behaviors in written communication, and they are developed in continuous dialogue with the practitioners who use them.
Everything in Sentinel and Dechefr is grounded in research, from the twelve indicators to the assessment workflow itself. Our team publishes in peer-reviewed journals on threat assessment and warning behaviors, and that work flows directly into the products.
Structured tools do not replace professional judgment. They support it. Sentinel and Dechefr are built to make your team's expertise more consistent, more transparent, and easier to defend.
For every team responsible for preventing targeted violence
Universities and schools
Campus threat assessment teams managing student-of-concern reports and communications.
Corporations
Corporate security and HR handling workplace violence prevention and insider concerns.
Public institutions
Agencies and authorities receiving threats and hostile communications directed at staff and officials.
Security and law enforcement
Professionals assessing threatening communications as part of protective operations.
See Sentinel and Dechefr on your own cases
Contact us for more information, and we'll walk your team through the full workflow, from intake to assessment to case management, using realistic material.
Contact usContact
Get in Touch
Ready to enhance your organization's security posture? Contact and Mind Intelligence Lab to discuss how our advanced insider threat detection and personnel vetting solutions can protect your critical assets.
Company
Mind Intelligence Lab