QG304 Clarification unbounded¶
| Type | Bug |
| Severity | High |
| Front end | a synthesised host and its router |
| Guardrail category (Vassilev 2026) | B. Policy & Governance Guardrails |
| AI RMF functions (NIST AI 100-1) | MAP, MANAGE |
| Why these functions | the catalogue's assignment, not the article's chain (GOVERN, MAP, MANAGE): an unbounded loop is a property of the router's structure, mapped and then managed by a bound; no governance rule is missing |
| OWASP | LLM06:2026 Unbounded Consumption |
| Gates by default | no, unless the project's quality profile adds it |
Raised when¶
Resolution or questioning can exceed the declared bound.
Why it matters¶
A clarification loop past its bound is a conversation that never ends or a question asked after the user was told the turn was over.
How to fix it¶
Declare the bound as a cited literal, and keep the generated router at the rung the bound was declared for.
What the finding carries¶
A location in the document the team edits, a fingerprint that survives unrelated edits, a reproduction (a transcript replayed against NeMo Guardrails for a policy, a dialogue case played through the selector for a router, or the gate's inputs reproduced by agreement), and, where the scan can produce one, a fix whose re-scan it has already run. See the catalogue for the other rules.