pj bee430cf72 fix(runner): no silent generator fallback, and llm on web ...

--generator llm against a spec declaring no generator = llm(...) logged a
warning and ran the seeded picker. For a comparison campaign that is silent
arm corruption: the run completes, the directory looks correct, and the wrong
policy drove it. It is now fatal.

pickSources also returned the V8 source for both action and extractor on web
before it looked at the generator, so the llm policy was unreachable there.
The two axes are now independent: the driver picks the extractor source, the
flag picks the action source, and llmSource composes with either because the
runner populates the candidate list and screenshot on every platform.

Claude-Session: https://claude.ai/code/session_01A5KmftdEJ49A9z5mF5ESrX

2026-08-12 18:56:08 +05:30

..

testdata

UX refactor (#52)

2026-06-02 09:52:53 +05:30

golden_test.go

UX refactor (#52)

2026-06-02 09:52:53 +05:30

llm_source_test.go

fix(runner): no silent generator fallback, and llm on web

2026-08-12 18:56:08 +05:30

llm_source.go

fix ltl semantics and unify action enumeration (#71)

2026-08-12 18:06:04 +05:30

runner_test.go

fix ltl semantics and unify action enumeration (#71)

2026-08-12 18:06:04 +05:30

runner.go

fix ltl semantics and unify action enumeration (#71)

2026-08-12 18:06:04 +05:30

source.go

fix(runner): no silent generator fallback, and llm on web

2026-08-12 18:56:08 +05:30

web_extractor_trace_test.go

fix ltl semantics and unify action enumeration (#71)

2026-08-12 18:06:04 +05:30