Commit graph

8 commits

Author SHA1 Message Date
Alessandro
daf95ec3ab Normalize tool contracts and slim prompt surface
Standardize multi-action tools around tool_args.action while keeping parser compatibility for older tool/args, tool_name:action, and method-shaped requests. This keeps new prompts clean without breaking agents that learned the previous dialect.

Move A0 connector remote execution/file tools into stable standard prompts, make remote targeting independent of the active chat context, and skill-gate beta computer-use remote so it no longer weighs down the always-on tool list.

Align text editor, scheduler, skills, office artifact, memory, notify, and browser prompts/tools around the canonical action contract. Add scheduler update/timezone handling, skills_tool read_file, text editor patch coverage, and fixes for memory_forget, behaviour_adjustment, and code execution progress warnings.

Reduce default prompt pressure by compacting browser and scheduler prompts into skill-backed manifests, shortening skill catalog descriptions, and pruning noisy framework knowledge. Remove obsolete connector prompt stubs and root tool-call knowledge examples.

Tests: conda run -n a0 pytest tests/test_a0_connector_prompt_gating.py tests/test_tool_action_contracts.py tests/test_task_scheduler_timezone.py tests/test_text_editor_context_patch.py tests/test_tool_request_normalization.py tests/test_office_document_store.py::test_odf_is_advertised_and_docx_remains_explicit_compatibility tests/test_office_document_store.py::test_document_artifact_accepts_method_alias_for_ods_create tests/test_skills_runtime.py tests/test_default_prompt_budget.py::test_a0_small_profile_removed_and_prompt_text_generic -q
2026-05-09 21:54:43 +02:00
Agent Zero Local
eecbb5ba34 fix(code_execution): handle closed PTYs while reading output 2026-05-02 21:33:08 +00:00
Agent Zero Local
a0f0c2e8d2 fix(code_execution): recover from closed PTY sessions
Detect closed or exited local TTY sessions before writing, convert invalid PTY write errors into retryable session failures, and reset/retry the terminal session once after send/read failures.
2026-05-02 21:32:12 +00:00
keyboardstaff
1bd5bc01d6 refactor(chat_branching): ID-based log ↔ history linking for precise branch trimming 2026-03-24 03:04:09 -07:00
frdel
6181ac9c20 refactor: mask secrets in code execution command output and increase truncation limit
- Import secrets helper module
- Increase initial command truncation from 200 to 250 characters
- Add secret masking to format_command_for_output using agent context's secrets manager
- Apply masking before final truncation to prevent secret exposure in logs
2026-03-22 16:57:51 +01:00
frdel
a82312f524 cache clearing, code exec heading trim 2026-03-19 20:50:37 +01:00
frdel
a7bfc7798b fix: OpenRouter header format, pipx install, code exec output truncation, and model config UI improvements
- Change X-OpenRouter-Categories from YAML list to comma-separated string
- Add pipx to base Python installation
- Truncate long lines in code execution output to 500 chars (250 from each end) before prompt detection
- Reorder model config UI sections and simplify plugin config button handler
- Remove border/padding from model summary grid and adjust mobile responsive styles
2026-03-19 16:28:33 +01:00
frdel
6515626242 refactor - plugin names and builtin plugins 2026-03-10 22:20:53 +01:00