- the fundamentals brief reached the model as a Python tuple
- analysts no longer emit a trade call that nothing reads
- a report that was not produced says so instead of appearing as a blank section
- PortfolioContext input, rendered once into state and read by the trader, risk and portfolio agents
- --portfolio JSON file on the CLI
- a run without it is never treated as a flat book
- the checkpoint signature keys on the portfolio
- the first speaker in each debate round received an empty opponent response,
yet the prompt demanded a rebuttal, so models fabricated the other side
- substitute an explicit opening marker when an opponent has not spoken, across
all five debators (bull, bear, and the three risk analysts) #1176
Agents had no ground-truth ticker→company mapping, so the market analyst
could pattern-match a price chart to the wrong company (e.g. TOTDY read as
"TotalEnergies"), and every downstream agent inherited the bad framing.
Resolve identity once at run start via a cached, fail-open yfinance lookup
and inject company/sector/exchange into the shared instrument context that
all twelve agents consume, with an explicit do-not-substitute instruction.
Resolution runs on both the propagate() and CLI entry points.
Also replaces the bare "Continue" message-clear placeholder, which some
OpenAI-compatible providers interpreted as the user task, with a
context-anchored placeholder carrying the resolved identity and date.
#814#888
output_language config now propagates to every user-facing agent.
Previously only the four analysts and portfolio manager respected
the setting, producing partial-localization reports with English
debate text interleaved with non-English analyst sections. Verified
live: 7 agents produce Chinese output when config is set to Chinese.
#575
