assess-australian-data-breach-l0cka · v0.1.0 · 2026-10-05 · sha256 e593b232fba982f0
assess-australian-data-breach-l0cka v0.1.0A
Immutable. This exact content never changes and is served at /api/v1/blob/e593b232fba982f0.
---
name: assess-australian-data-breach-l0cka
title: Assess Australian Data Breach
description: Triage a suspected Australian privacy or cybersecurity incident for lawyers; extract incident facts; identify potentially applicable Commonwealth, State, Territory and sectoral breach regimes; perform change-safe Australian Privacy Principles verification where an APP issue is material; verify Commonwealth and NSW legislation through the corresponding legislation skills; distinguish statutory, regulatory, contractual and voluntary reporting; and prepare a time-sensitive assessment and action matrix. Use for suspected unauthorised access, disclosure or loss, ransomware, cyber extortion, vendor incidents and breach-response planning. Do not use for case law, forensic conclusions, live containment, threat hunting, external reports or autonomous notification decisions.
author: l0cka
author_url: https://github.com/l0cka/legal-skills/tree/main/plugins/australian-privacy-cybersecurity/skills/assess-australian-data-breach
license: MIT
version: 0.1.0
execution_mode: open
jurisdiction: au
practice: data-protection
language: en
sources:
- title: Breach Assessment Method
path: references/breach-assessment-method.md
---
# Assess Australian Data Breach
Prepare a privileged-workstream-ready preliminary assessment for a lawyer. Do
not conclude that a breach is notifiable merely because a security incident
occurred, and do not delay urgent human escalation while researching.
Read [references/breach-assessment-method.md](references/breach-assessment-method.md)
before starting. It defines the fact schema, regime prompts, deadline controls
and source rules.
## Workflow
1. Escalate and protect.
- Flag immediate safety, ongoing compromise, destructive activity, material
service disruption or an apparently running statutory clock for urgent
human attention.
- Ask only for the minimum de-identified facts. Do not request credentials,
live malware, exploit code, unnecessary identifiers or complete datasets.
- Do not take containment action, contact a threat actor, make a payment,
notify a person or regulator, or submit a report. Preserve the lawyer's
approval boundary and the forensic evidence-preservation plan.
2. Establish the incident record.
- Separate known facts, source of each fact, estimates, inferences,
assumptions, disputed facts and unknowns.
- Record detection and awareness times with timezone; suspected occurrence
window; systems and custodians; data types and protections; affected and
potentially affected people; location; actors; containment/remediation;
recipients; vendors; critical services; ransom demand or payment; and
existing notices.
- Do not equate detection, confirmation, containment, assessment completion
or occurrence. Record each timestamp independently.
3. Map candidate regimes.
- Test the Privacy Act Notifiable Data Breaches scheme and each applicable
State, Territory, health, critical-infrastructure, cyber-extortion,
telecommunications, financial-services or contractual regime.
- Distinguish mandatory notice, assessment duties, voluntary reports,
contractual notice, insurance conditions and internal escalation.
- Treat numerical thresholds, triggering knowledge standards, recipients,
content requirements and time periods as propositions requiring current
verification.
- Invoke `$route-australian-privacy-jurisdiction` for every State or
Territory nexus and `$map-australian-cyber-incident-obligations` for the
concurrent Commonwealth and sectoral matrix.
4. Verify law and calculate cautiously.
- Invoke `$check-commonwealth-legislation` for every material Commonwealth
title or provision and the relevant incident date. Invoke
`$check-nsw-legislation` for every material NSW title or provision.
- Invoke `$check-australian-privacy-principles` when security, retention,
use, disclosure or another APP issue is material. Keep this Schedule 1
check separate from verification of the Notifiable Data Breaches scheme.
A detected or unverified APP-framework change blocks the APP conclusion,
but must not delay urgent containment or independently verified incident
duties.
- Invoke `$assess-statutory-privacy-tort` independently of APP coverage when
intentional or reckless intrusion, surveillance, misuse, disclosure or
publication may be material.
- For another State or Territory, inspect its official legislation source.
If a version or commencement point cannot be established, use `NOT
VERIFIED` and escalate the check.
- Calculate a deadline only after verifying the trigger, period, counting
rule and relevant facts. Show the trigger time, timezone, calculation and
uncertainty. Never invent a date from an incomplete record.
- State that case law was not considered. Do not search for or rely on it.
5. Assess and hand off.
- Address reasonable-likelihood, serious-harm, remedial-action and
multi-entity issues only against verified text and stated facts.
- Build one row per possible obligation. Give the earliest plausible
deadline prominence without describing an uncertain date as final.
- Identify the decision owner, evidence needed and next action. Reserve all
notification and legal conclusions for a lawyer and authorised incident
lead.
## Result contract
Lead with `URGENT PRELIMINARY BREACH ASSESSMENT — HUMAN DECISION REQUIRED`, then
provide:
```text
Incident status: <suspected/ongoing/contained/unknown, not a legal conclusion>
Scope and exclusions: <including no case law and no forensic opinion>
Jurisdictions and sectors: <identified and unresolved>
As at: <YYYY-MM-DD HH:MM timezone>
Immediate human escalations: <owner and reason>
Known facts, assumptions, disputes and unknowns: <separate sections>
APP framework status: <exact result and effect on APP analysis, if relevant>
Potential-obligation matrix:
Priority | Regime | Entity/role | Trigger and threshold | Trigger facts |
Verification status | Earliest plausible deadline | Recipient/action |
Source | Missing evidence | Decision owner
Harm and remediation assessment: <facts for and against, uncertainty retained>
Notification coordination: <statutory, contractual, insurer and voluntary channels separated>
Concurrent cyber map: <one row and one clock per entity and regime>
Statutory privacy tort: <separate screen or reason not triggered>
Action plan: <owner, action, time, dependency and approval>
Preservation and privilege considerations: <questions for the lawyer and forensic lead>
Limitations and review: <currency, factual and interpretation gaps>
```
Use `VERIFIED`, `VERIFIED WITH QUALIFICATIONS`, `NOT VERIFIED`, `PRIMARY TEXT
CHECK REQUIRED`, or `OUTSIDE SCOPE` for legislation status. Link the official
text and version actually inspected. Do not describe an OAIC or ACSC web form,
guideline, policy, standard or contract as legislation.
## Example
An abbreviated completed assessment for a vendor ransomware incident. Every
value is illustrative, and a real assessment includes one matrix row per
possible obligation, not the two shown.
```text
URGENT PRELIMINARY BREACH ASSESSMENT — HUMAN DECISION REQUIRED
Incident status: Suspected, containment unconfirmed (not a legal conclusion)
Scope and exclusions: No case law considered; no forensic opinion given
Jurisdictions and sectors: Commonwealth confirmed; NSW health nexus unresolved
As at: 2026-08-14 09:30 Australia/Sydney
Immediate human escalations: Incident lead — ransom demand received and a
statutory assessment clock may already be running
Known facts, assumptions, disputes and unknowns: Detection 2026-08-13 22:10
AEST (known, monitoring alert); occurrence window unknown; exfiltration
assumed but disputed by vendor
APP framework status: VERIFIED WITH QUALIFICATIONS — APP 11 verified against
the current compilation; a commenced-but-unincorporated amendment was
flagged and blocks an unqualified APP conclusion
Potential-obligation matrix (abbreviated):
1 | Privacy Act NDB scheme | APP entity | Reasonable grounds to suspect an
eligible data breach → 30-day assessment | Detection facts above |
VERIFIED | 2026-09-12 (from detection; trigger time uncertain) | OAIC +
affected individuals | <official compilation link> | Exfiltration evidence |
General counsel
2 | Vendor MSA cl 14 notice | Customer of vendor | "Security incident" defined
term, 48 hours | Vendor notice 2026-08-14 07:00 | OUTSIDE SCOPE (contract,
not legislation) | 2026-08-16 07:00 | Customer security contact | MSA v3.2 |
Executed MSA copy | Commercial lead
Harm and remediation assessment: Facts for and against serious harm retained
separately; no conclusion drawn
Action plan: Preserve logs (forensic lead, today); confirm data classes
(vendor, 48h); lawyer decision on NDB assessment start time
Limitations and review: Occurrence window, exfiltration and NSW nexus
unverified; all notification decisions reserved to the lawyer
```
<!-- generated:plugin-dependencies -->
## Other plugins
This skill names skills from other Legal Skills plugins. Install a required
plugin alongside this one; install a hand-off plugin when the matter needs
that depth.
- `australian-legal-research` (required): `check-commonwealth-legislation`,
`check-nsw-legislation`. Verifies the Commonwealth, state and territory
legislation the privacy, breach and cyber workflows rely on.
If a named skill is not available in this session, say so and name the plugin
to install. Treat the step it would have performed as not done: record the
affected proposition as not verified, or the hand-off as not made. Never
perform that step from memory or substitute a different skill.
<!-- end:plugin-dependencies -->