Threat intelligence reports fail when readers cannot tell in thirty seconds what happened, how sure the analyst is, who they may share it with and what they should do next. This template gives [Organisation]'s threat intelligence team one consistent structure for flash alerts, incident-driven reports and periodic assessments: a bottom line up front (BLUF), calibrated likelihood and confidence language from US ICD 203, FIRST Traffic Light Protocol (TLP) 2.0 markings, and recommended actions grouped by response phase so the SOC and incident responders can act on them directly. It is written for CTI analysts, SOC leads and incident response managers who produce or consume intelligence. For the wider discipline, see the Threat Intelligence Guide; replace every [placeholder] before use.
How to use this template
- Pick the report type first (section 1.2). The structure stays the same; the length and audience change.
- Write the BLUF last, place it first. It must stand alone if the reader stops there.
- Use only the likelihood terms in section 3.3 and state confidence separately. Never invent new probability words.
- Mark every report with TLP in the header and footer of each page, and in the email subject line when sent by email.
- Keep the action sections (4 to 6) concrete: each action names an owner team, a system and a due time.
- Link it to requirements. Every report answers at least one intelligence requirement. See the Intelligence Requirements Template.
1. Scope and triggers
1.1 This template applies to all finished intelligence produced by [Organisation]'s CTI function for internal teams, executives, customers or sharing partners.
1.2 Report types and triggers:
| Type | Trigger | Audience | Target length | Due |
|---|---|---|---|---|
| Flash alert | Active exploitation of a technology [Organisation] runs; credible targeting of [Organisation] or its sector; leaked [Organisation] credentials | SOC, IR, system owners | 1 page | [2] hours from validation |
| Incident intelligence report | A declared incident needs actor, technique or infrastructure context | Incident lead, IR team | 2 to 4 pages | [24] hours, then updates |
| Threat actor or campaign assessment | A priority intelligence requirement, new campaign against the sector | Security leadership, detection engineering | 4 to 8 pages | [5] working days |
| Strategic assessment | Quarterly cycle, board request, major geopolitical change | Executives, risk committee | 2 to 6 pages, non-technical | [Quarterly] |
2. Roles and contacts
| Role | Named lead | Responsibilities |
|---|---|---|
| Author (CTI analyst) | [Name] | Collects and grades sources, writes judgments, proposes TLP and actions |
| Peer reviewer | [Name] | Challenges judgments and alternatives, checks indicators and technique IDs |
| Approver (CTI lead) | [Name] | Approves release, confirms TLP and distribution list |
| Action owners | [SOC lead, IR lead, vulnerability management lead] | Accept, reject or schedule each recommended action and report status |
| Sharing liaison | [Name] | Releases sanitised versions to ISACs and partners within the original TLP limits |
2.1 Flash alerts may be released by the author with verbal approval from the CTI lead, recorded afterwards. All other types need written approval.
3. Detection and triage
3.1 Source grading. Grade each source and each piece of information separately, using the Admiralty scale: source reliability A (completely reliable) to F (cannot be judged), and information credibility 1 (confirmed by other sources) to 6 (cannot be judged). Record the grade beside each source in the sources section (7.2, part H). See the Intelligence Confidence and Source Reliability Guide.
3.2 Triage questions before writing: Does this affect technology, identities or suppliers that [Organisation] actually uses? Is there evidence of exploitation or targeting, or only a proof of concept? What would the reader do differently because of this report? If the answer to the last question is nothing, do not publish a report; log the item instead.
3.3 Likelihood language. Use only these terms, taken from ICD 203, and do not mix terms from different rows:
| Term (use one pair) | Approximate probability |
|---|---|
| Almost no chance / remote | 1 to 5% |
| Very unlikely / highly improbable | 5 to 20% |
| Unlikely / improbable | 20 to 45% |
| Roughly even chance / roughly even odds | 45 to 55% |
| Likely / probable | 55 to 80% |
| Very likely / highly probable | 80 to 95% |
| Almost certain / nearly certain | 95 to 99% |
3.4 Confidence. State confidence (high, moderate or low) in a separate sentence from the likelihood statement, as ICD 203 advises, and base it on the quality of the evidence and the logic behind the judgment. Example: "We assess it is likely that the group will target [sector] VPN appliances in [quarter]. We have moderate confidence in this judgment, based on two reliable sources and one partially corroborated report."
3.5 Separate fact from judgment. Facts are written as facts ("The domain was registered on [date]"); judgments always begin with "We assess" or "We judge". This follows the ICD 203 standard that analysis must distinguish underlying information from analysts' assumptions and judgments.
4. Containment
4.1 Report section E, "Immediate actions", lists what to do now to stop or limit harm. Each line has the form: [Action] on [system or control] by [owner] within [time].
- Block: indicators to block, with type, value, first seen, source grade and expiry date (indicators age quickly; set an expiry of [30] days unless renewed).
- Exposure check: the query or asset search that tells owners whether they run the affected product or version.
- Credential action: where the report concerns stolen credentials or tokens, which accounts, keys or sessions to revoke. See the Threat Intelligence for Identity and NHI Guide.
5. Eradication
5.1 Report section F, "Hunting and detection", gives what defenders need to find and remove existing activity:
- Techniques: MITRE ATT&CK technique IDs observed, for example T1566.002 Spearphishing Link, T1190 Exploit Public-Facing Application or T1078 Valid Accounts, with the specific procedure seen.
- Hunt hypotheses: one to three testable statements with the data source to search.
- Detection content: links to Sigma or SIEM rules in the detection repository, not pasted queries that drift out of date. See the Sigma Rules Guide.
6. Recovery
6.1 Report section G, "Longer-term mitigations", covers controls that reduce exposure after the immediate threat passes: patching priorities, configuration changes, architectural changes and the risk register entry to raise. Each item names the owner and whether it is accepted, scheduled or rejected (status is tracked in [ticketing system], not in the report).
7. Communication and notification
7.1 TLP 2.0 marking. FIRST TLP 2.0 has been the current version since August 2022. Write labels in capitals without spaces, put them in the header and footer of every page, and in the subject line and immediately before the information in emails.
| Label | Who recipients may share with | Typical use at [Organisation] |
|---|---|---|
| TLP:RED | Named individual recipients only | Active incident details, named victims |
| TLP:AMBER+STRICT | Recipients' organisation only | Internal reports containing [Organisation] system details |
| TLP:AMBER | Recipients' organisation and its clients, need to know | Reports shared with managed service providers |
| TLP:GREEN | Recipients' community, not public | Sanitised sharing with an ISAC |
| TLP:CLEAR | No limit | Public advisories, published blog content |
See the Traffic Light Protocol Guide.
7.2 Report skeleton (keep the lettered parts in this order):
- Header: TLP label, report ID, type, date and time (UTC), version, author, approver, intelligence requirement answered.
- BLUF: two to four sentences: what is happening, why it matters to [Organisation], the key judgment with likelihood, and the single most important action.
- Key judgments: numbered, each with likelihood (3.3) and a separate confidence sentence (3.4).
- Details: facts, timeline, actor and technique context, relevance to [Organisation] (affected assets, business services).
- Immediate actions (section 4).
- Hunting and detection (section 5).
- Longer-term mitigations (section 6).
- Sources and grading: each source with its Admiralty grade; alternative explanations considered and why they were rejected; intelligence gaps.
- Indicator annex: machine-readable file (STIX 2.1 bundle or CSV) attached, not embedded in prose.
- Feedback: link or address for readers to report relevance and actions taken.
Worked example of parts B and C (fictional):
BLUF: A criminal group is exploiting an authentication bypass in [VPN product] version [x], which [Organisation] runs at [number] remote access gateways. We assess it is very likely that unpatched [Organisation] gateways will be targeted within [7] days. The network team should apply the vendor fix or disable the affected portal by [date, time UTC].
Key judgment 1: We assess it is likely that the group uses access to steal VPN session tokens and reuse them from hosting provider addresses. We have moderate confidence in this judgment: it rests on one vendor advisory (graded B2) and one ISAC report (graded C3) that describe the same procedure.
Gap: We have no reporting on whether patched gateways that were exploited before patching keep attacker sessions. We will update this report when the vendor answers.
Notice what the example does: it names [Organisation]'s exposure, gives a deadline, keeps likelihood and confidence in separate sentences, shows source grades and states what is not known. A reader who stops after the BLUF still knows the one action to take.
7.3 Machine-readable version. Where the report is shared through a threat intelligence platform, publish it as a STIX 2.1 Report object with its indicators and attack patterns, the TLP marking definition applied, and the STIX confidence property (0 to 100) mapped from the confidence level. See the STIX and TAXII Guide.
7.4 Updates. Issue a new version, not an edit in place. State at the top what changed and whether any judgment changed, as ICD 203 requires analysts to explain change to or consistency of judgments.
8. Post-incident review
8.1 For each incident intelligence report and each quarter of flash alerts, review: were the key judgments accurate, were actions taken and on time, which sources proved reliable, and what feedback readers gave.
8.2 Track these measures: time from validation to flash alert release; percentage of recommended actions accepted and completed on time; percentage of reports with reader feedback; detection rules created from reports; and judgments later confirmed or refuted.
8.3 Re-grade sources whose reporting proved wrong and record lessons in the CTI style guide.
Implementation guidance
- Build the template in your platform. Create the lettered skeleton as a document template and as a report object template in the threat intelligence platform, with the TLP header and footer pre-filled.
- Train the language. Run a short calibration session where analysts rate sample statements using the likelihood table, until terms are used consistently.
- Agree action routing with the SOC, IR and vulnerability management leads, including how actions arrive as tickets and the response times they will meet.
- Pilot with the SOC for [one month] using flash alerts only, collect feedback, then extend to the other types.
- Map to frameworks: the process supports NIST CSF 2.0 ID.RA-02 (threat intelligence received from sharing forums and sources), DE.AE-07 (threat intelligence integrated into analysis) and RS.CO-03 (information shared with stakeholders), and NIST SP 800-53 Rev. 5 PM-16 Threat Awareness Program, SI-5 Security Alerts, Advisories and Directives, and RA-10 Threat Hunting.
Common failure patterns to avoid: a BLUF that summarises the news instead of stating the impact on [Organisation]; likelihood and confidence blended into phrases such as "highly likely, high confidence"; indicator lists without dates or expiry; TLP in the body only, so it is lost when pages are forwarded; and recommendations with no owner, which nobody acts on.
Related NHI Mgmt Group resources: Threat Intelligence Guide · Intelligence Requirements Template · Traffic Light Protocol Guide · Threat Intelligence Maturity Model