Assumption Registry — {DOMAIN_DESCRIPTION}

Domain Assumptions

Known assumptions that reasoning in this domain depends on. Every assumption has a sensitivity rating and, where possible, a concrete test method.

ID	Assumption	Type	Sensitivity	Testable	Test Method
AS-001	{ASSUMPTION_STATEMENT}	{TYPE}	{critical/high/moderate/low}	{yes/no}	{HOW_TO_VERIFY}
AS-002	{ASSUMPTION_STATEMENT}	{TYPE}	{critical/high/moderate/low}	{yes/no}	{HOW_TO_VERIFY}
AS-003	{ASSUMPTION_STATEMENT}	{TYPE}	{critical/high/moderate/low}	{yes/no}	{HOW_TO_VERIFY}

Assumption Types

Type	Description	Examples
access	Permissions, credentials, reachability	“Service has read access to database”
state	Current condition of systems or data	“Table contains records from last 30 days”
behavioral	Expected behavior of systems or people	“API returns JSON, not XML”
methodological	Analytical or procedural choices	“Linear relationship between X and Y”
preference	Stakeholder preferences not yet confirmed	“Client prefers dashboard over PDF report”

Sensitivity Ratings

Rating	Meaning	Required Action
critical	If violated, ALL dependent work is invalid	MUST test before proceeding
high	If violated, THIS TASK fails	Should test before proceeding
moderate	If violated, results may be degraded	Test when convenient
low	If violated, minor impact	Document but no blocking test needed

---

Analytical Assumptions (Per-Task)

For every analysis or implementation task, enumerate assumptions in this format:

ID	Assumption	Type	Sensitivity	Status
[auto]	[Statement]	[Type from above]	[Rating]	untested / confirmed / violated

---

Adding New Assumptions

When a new assumption is discovered during work:

1. Assign next available ID in the appropriate category
2. Rate sensitivity using the scale above
3. Determine testability and write a concrete test method if possible
4. If sensitivity: critical, test BEFORE proceeding with dependent work
5. If assumption is violated, update status and halt dependent work (see Halt on Contradiction protocol)