haiku.rag/haiku_rag_slim/haiku/rag/capabilities
Yiorgis Gozadinos bae359e3bf
Align the MCP guidance with the analysis instructions and move to Monty 0.0.23
The execute_code description gains the toc.json node shape, the patterns
the evals put into the analysis instructions (one list_documents call to
map a title to an id, files carry no source, toc before search for a known
document, doc_item_refs are items self_refs, chunk ids join files and are
not citations) and the sandbox's read-only, no-network, time-limit and
output facts, which the analysis instructions now state too. The skill
gains pictures as images, image search when offered, and citing chunk
metadata locators. docs/mcp.md lists the interpreter's limits under Code.

pydantic-monty>=0.0.23 brings collections, itertools, functools,
dataclasses, function decorators and str.format into the sandbox; every
layer names the same modules, and a test imports them. Monty now caps host
callbacks per checkout at 1000 by default; the sandbox raises it out of
reach, since the time budgets govern.
2026-09-07 12:02:33 +03:00
..
instructions Align the MCP guidance with the analysis instructions and move to Monty 0.0.23 2026-09-07 12:02:33 +03:00
__init__.py Give the evidence capabilities one typed state 2026-08-19 17:08:44 +03:00
_base.py Remove the environment overrides and the last unnamed-database wording 2026-09-03 15:12:09 +03:00
_tools.py Deduplicate search results within one model response 2026-09-02 12:29:12 +03:00
analysis.py Replace ask_question and analyze with execute_code 2026-09-07 10:02:42 +03:00
compaction.py Name the collection on a retrieved image, not only in the text 2026-08-28 12:02:16 +03:00
evidence.py Give the evidence capabilities one typed state 2026-08-19 17:08:44 +03:00
ledger.py Let the record say whether a question is still being answered 2026-08-13 15:03:24 +03:00
policy.py Support Pydantic AI agent specs via from_spec 2026-08-21 13:04:09 +03:00
rag.py Price qa.max_searches in search units 2026-09-02 12:29:12 +03:00