A quick search on the internet confirms that "黄芪" is also known as "芰草", therefore it is extremely likely that the proposed character is an error form for 芰. In this light, the evidence shown is not sufficient for encoding, and the character should be withdrawn.

I agree with Andrew's comment.

Agree with Andrew and Lee. If there is no other sufficient evidence, this character should not be accepted.

It is also SAT-04688.

唐駢體文鈔

十萬卷樓叢書·釋文

[艹攴] might be regarded as the error form of 芰, but it is used in different sources. Could we confirm whether it can be accepted as having sufficient evidence of use?

The new evidences come from 唐駢體文鈔 and 十萬卷樓叢書·釋文, but when looking for other version, most of them are given as 芰.
1. 唐駢體文鈔(光緒乙未刊行) is [芰]. And the poem included in this book (蜀綿州月王樓詩序) are given as [芰] in 御定全唐詩.
唐駢體文鈔(光緒乙未刊行)


2. In 十萬卷樓叢書, it wrote “竒寄切。說文，[艹攴] 蔆也.” However, “竒寄切” is the reading of 支, not [攴].