For Gemini's implicit cache (and similar provider caches) to actually
hit step over step, the rendered transcript for steps 1..N-1 must be
byte-identical at step N and at step N+1. The agent already appends
HistoryItems immutably in practice, but nothing in the type prevented
a future caller from mutating one in place — which would silently kill
the cache from that byte onward.
Mark HistoryItem frozen so any mutation now fails loud at runtime
rather than slowly burning input tokens. Add a regression test set
that asserts the cache property directly:
- render(items[:N]) must be a strict byte-prefix of render(items[:N+1])
- to_string() is deterministic for identical inputs
- the prefix property holds across mixed entry shapes (normal steps,
errors, system messages, follow-up tasks)
- conditional field inclusion doesn't collapse to ambiguous output
Known limitation, not addressed here: max_history_items compaction in
MessageManager.agent_history_description rewrites earlier bytes once
the cap is exceeded (the omitted-count message changes). That's a
larger redesign and deserves its own PR.
- Add max_clickable_elements_length parameter to Agent.__init__ with default 40000
- Add max_clickable_elements_length to AgentSettings
- Pass parameter through MessageManager to AgentMessagePrompt
- Bump version to 0.11.10a2
Co-authored-by: Magnus Müller <MagMueller@users.noreply.github.com>
Fixes#3970
The LLM was typing literal placeholder names (e.g., 'user_name') instead
of wrapping them in <secret> tags (e.g., '<secret>user_name</secret>').
Changes to _get_sensitive_data_description():
- Added clear 'SENSITIVE DATA' header to draw attention
- Changed placeholder list format from Python list to bullet points
- Added explicit 'MUST' instruction emphasizing the required format
- Added concrete example using the first actual placeholder name
- Added explanation that the system will replace tags with actual values
This makes it much clearer to the LLM that it needs to wrap placeholder
names in <secret> tags for the automatic replacement mechanism to work.
This change ensures that sensitive data in action results is filtered before being included in state messages sent to the LLM. This prevents accidental leakage of private information. New tests are added to verify this functionality.
Co-authored-by: mailmertunsal <mailmertunsal@gmail.com>
- Updated page info text format to enhance readability and structure.
- Changed date format in agent state description to display only the current date.
- Simplified action results formatting in message manager for better clarity.
- Adjusted history item string representation to streamline output and remove unnecessary prefixes.
MagMueller