You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
fix(tools): support legacy strands module-based tool pattern
- Update tool loader to fall back to strands module pattern when no @tool-decorated functions found
- Import strands load_tools_from_module for TOOL_SPEC-based tool compatibility
- Add test coverage for TOOL_SPEC module pattern, private tools, and error cases
Copy file name to clipboardExpand all lines: docs/configuration/Chapter_05.md
+2-2Lines changed: 2 additions & 2 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -16,7 +16,7 @@ agents:
16
16
- ./utils/ # All @tool functions from all .py files in dir
17
17
- my_package.tools # All @tool functions from an installed module
18
18
- my_package.tools:special_function # One specific function from a module
19
-
- strands_tools:http_request # A tool from strands' built-in tools
19
+
- strands_tools.http_request # A tool from strands' built-in tools
20
20
system_prompt: "You analyze text using your tools."
21
21
```
22
22
@@ -97,7 +97,7 @@ tools/
97
97
> **Tips & Tricks**
98
98
>
99
99
> - Organize tools in a directory when you have many of them. One file per domain: `tools/math.py`, `tools/text.py`, `tools/database.py`.
100
-
> - The `strands_tools` package has built-in tools like `http_request`, `file_read`, `shell` — use them with `strands_tools:http_request`.
100
+
> - The `strands_tools` package has built-in tools like `http_request`, `file_read`, `shell` — use them with `strands_tools.http_request`.
101
101
> - Each agent gets its own copy of tools. Two agents referencing the same file get independent tool instances.
102
102
> - Tool function docstrings are sent to the LLM as the tool description. Write good docstrings — they directly affect how well the model uses your tools.
103
103
> - Type hints on tool parameters become the JSON schema the LLM sees. Use `str`, `int`, `float`, `bool`, `list[str]`, etc. The more specific your types, the better the LLM calls your tools.
0 commit comments