Reviewed 18 August 2026 · next review 17 February 2027
Safety warning wording
A warning that names only the hazard leaves the reader without the two things they need: what happens, and what to do instead.
Last updated · Written and reviewed by Jonatan Tensetti · Tensetti Tools
ClearSpec uses an independent proprietary clarity method. It is not affiliated with ASD/STEMG, is not an ASD-STE100 checker or certification service, and does not certify ISO 24495-1 conformance.
Three parts, every time
A usable warning states the hazard, the consequence of ignoring it, and the avoidance action. The check looks for all three after a DANGER, WARNING, CAUTION or NOTICE signal word and reports which part it could not find.
- CS_WARNING_SEQUENCE — Safety-language review recommended: this warning statement does not clearly state the hazard and the consequence and how to avoid it. State the hazard, the consequence and the avoidance action, then have a competent person approve the wording.
Possible false positive: The missing part may appear in an adjacent block or a graphic.
- No finding from these checks in the current wording.
Language ClearSpec deliberately avoids
- It writes safety-language review recommended, never “unsafe” and never “compliant”.
- It never generates replacement safety wording.
- It flags the structure only; the hazard analysis behind the wording is out of scope.
Related checks in safety documents
When you select the safety document type, requirement sentences with no actor and passive instructions are raised to high severity, and weak obligation words such as should are treated as high rather than medium.
- CS_WEAK_MODAL — 'should' does not state whether the step is mandatory or optional. Use 'must' for an obligation, or make the optional nature explicit.
- CS_PASSIVE_ACTION — The action is written in the passive voice and no actor is named. Name the actor and use an active instruction, for example 'The technician closes the valve.'
Possible false positive: Genuinely optional advice should use a permissive word.
22 checks, no account, and the text is analysed in your browser.
Questions
- Does ClearSpec decide whether my warning is adequate?
- No. It reports that a structural part appears to be missing and recommends a safety-language review. Adequacy is decided by a competent person against the standards that apply to your product.
- Why does it never say a sentence is unsafe?
- An automated heuristic cannot establish safety. The finding is phrased as safety-language review recommended so nobody mistakes it for an approval or a rejection.
- Will it rewrite safety text automatically?
- Never. ClearSpec does no generative rewriting, and safety instructions are explicitly excluded from any future automatic rewrite.
References
Checked 18 August 2026. Referenced for context only; no standard text or dictionary content is reproduced.
- Clear content objective — W3C WAI. Accessibility-oriented clear-content guidance informing general principles.
- Writing for web accessibility — W3C WAI. First-party accessibility writing tips informing general principles.
ClearSpec reports automated heuristics from method version 1.0.0. The clarity indicator is not a measure of technical accuracy, usability, legal sufficiency, safety or standards conformance. A competent person must decide the final wording, especially for safety content.