First, hey Peter, thanks for input! I'd seen the explicit piping before, couldn't remember its format though; I will switch to it. Well, after a morning of testing, I see the test system is actually finding the repeat section variables. It looks like its error "Parameter filter_column requires a value, but has no legal values defined" points to the fact that it can't process any option list having <options from_data_table="[whatever table we got]"> . I'd assumed Planemo in its test setup would install the tool's data tables, but now I see it is still on John's enhancement list. So I'll table those tests until that's a go. I guess manually setting up the data tables in galaxy won't solve that for planemo? ************************* I'm still stuck trying to get tool test directly in a fresh copy of Galaxy. I edit config/tool_conf.xml.sample to have <section id="testing" name="TESTING"> <tool file="TESTING/blast_reporting/blast_reporting.xml" /> </section> And I run "tests.sh -id blast_reporting" (with Galaxy as owner of tool folder): ====================================================================== ERROR: Failure: ValueError (No such test TestForTool_blast_reporting) ---------------------------------------------------------------------- Traceback (most recent call last): File "/usr/local/galaxy/production2/galaxy-dist/eggs/nose-0.11.1-py2.6.egg/nose/failure.py", line 39, in runTest raise self.exc_class(self.exc_val) ValueError: No such test TestForTool_blast_reporting Though planemo has no problem linting and testing it. Arg! ************************* Thanks, Damion d. p.s. I did mention first test works (in p1) :) .