RomanNum3ral
/
cai
mirror of https://github.com/aliasrobotics/cai.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
236 Commits 6 Branches 5 Tags 218 MiB
Commit Graph

48 Commits

Author SHA1 Message Date
Rohan Mehta 68c800d2a3 [2/n] Add MCP support to Runner 
### Summary:
This enables users to **use** MCP inside the SDK.
1. You add a list of MCP servers to `Agent`, via `mcp_server=[...]`
2. When an agent runs, we look up its MCP tools and add them to the list of tools.
3. When a tool call occurs, we call the relevant MCP server.

Notes:
1. There's some refactoring to make sure we send the full list of tools to the Runner/Model etc.
2. Right now, you could have a locally defined tool that conflicts with an MCP defined tool. I didn't add errors for that, will do in a followup.

### Test Plan:
See unit tests. Also has an end to end example next PR.
 2025-03-24 15:08:02 -04:00
Rohan Mehta 300e12c198 [1/n] Add MCP types to the SDK 
### Summary:
1. Add the MCP dep for python 3.10, since it doesn't support 3.9 and below
2. Create MCPServer, which is the agents SDK representation of an MCP server
3. Create implementations for HTTP-SSE and StdIO servers, directly copying the [MCP SDK example](https://github.com/modelcontextprotocol/python-sdk/blob/main/examples/clients/simple-chatbot/mcp_simple_chatbot/main.py)
4. Add a util to transform MCP tools into Agent SDK tools

Note: I added optional caching support to the servers. That way, if you happen to know a server's tools don't change, you can just cache them.

### Test Plan:

Checks pass. I added tests at the end of the stack.
 2025-03-24 15:08:02 -04:00
Rohan Mehta c0794a90ec Read tracing API data lazily 2025-03-21 15:32:37 -04:00
Richie Caputo b5305810d7
Create py.typed 
- Ensure library is properly typehinted
 2025-03-21 11:16:24 -04:00
Kazuhiro Sera 98c4b45b6a Make the optional dependency installation compatible with zsh 2025-03-21 10:01:24 +09:00
Rohan Mehta 1771c1e856 update tests 2025-03-20 13:08:38 -04:00
Dominik Kundel c7ce154637 feat: add voice pipeline support 
> Co-authored-by: rm@openai.com
 2025-03-20 09:43:13 -07:00
Rohan Mehta 8c9974bc90 Fix breaking changes from openai 1.66.2 2025-03-19 12:30:45 -04:00
Rohan Mehta 10aa5555af Introduce tool_use_behavior on agents 2025-03-18 21:55:12 -04:00
Rohan Mehta 6f7e801da0
Update tests and docs for strict mode decorator (#205) 
As titled. Test plan: unit tests/docs.
 2025-03-18 21:35:00 -04:00
Rohan Mehta d0a7b00086
Fix typo on Agent documentation (#151) 
Argument name is not description but handoff_description
 2025-03-17 16:51:47 -04:00
Carlos Souza 64150c8aea Fix lint 2025-03-17 16:14:28 -04:00
Rohan Mehta 0989cd9184
Add TracingProcessor export to __init__.py (#192) 
# Summary
This adds the missing TracingProcessor export to  __init__.py.

# Behavior
When trying to add a custom tracing processor, the TracingProcessor
importing fails with not found error when trying the example usage
proposed in issue #164

Specifically this line throws the error:
`add_trace_processor(MyTracingProcessor("output"))`

# Expected Behavior
Inspecting the init file, simply the import/export was missing. Adding
these made the example code work for me

# Test plan
Local dev of example code in #164

# Issue number
#164

# Checks
None
 2025-03-17 15:39:04 -04:00
Rohan Mehta 47aed7d362 Update tests and docs for strict mode decorator 2025-03-17 15:06:57 -04:00
Rohan Mehta 64e263b614 Pretty print result classes 2025-03-17 11:11:39 -04:00
Vincenzo Domina 1f58528f1c
Add TracingProcessor export to __init__.py 2025-03-17 10:37:43 +01:00
Rohan Mehta 09d70c074d utils directory 2025-03-16 18:48:45 -04:00
Rohan Mehta 951193bd21
feat: Add strict_mode option to function_schema and function_tool (#60) 
This PR introduces a `strict_mode: bool = True` option to
`@function_tool`, allowing optional parameters when set to False. This
change enables more flexibility while maintaining strict JSON schema
validation by default.

resolves #43 

## Changes:

- Added `strict_mode` parameter to `@function_tool` and passed it to
`function_schema` and `FunctionTool`.
- Updated `function_schema.py` to respect `strict_mode` and allow
optional parameters when set to False.
- Added unit tests to verify optional parameters work correctly,
including multiple optional params with different types.

## Tests:

- Verified function calls with missing optional parameters behave as
expected.
- Added async tests to validate behavior under different configurations.
 2025-03-16 17:43:46 -04:00
Carlos Souza 26828e5e68
Fix typo on Agent documentation 
Argument name is not description but handoff_description
 2025-03-13 16:18:40 -04:00
Rohan Mehta 7f68a7da2a
Consolidate to one logger (#149) 
Results in double logging if we dont. Closes #141.
 2025-03-13 11:38:18 -07:00
Rohan Mehta 17f0a425ba Consolidate to one logger 2025-03-13 13:55:15 -04:00
Rohan Mehta 4db24bdb3c Update tracing docs to be correct 2025-03-13 13:20:27 -04:00
Rohan Mehta 16cfce117c
Add examples and documentation for using custom model providers (#110) 2025-03-13 08:17:25 -07:00
Rohan Mehta 341198ff92 Add request ID 2025-03-12 18:07:59 -07:00
Rohan Mehta 25a633139d Add examples and documentation for using custom model providers 2025-03-12 17:34:38 -07:00
Rohan Mehta 2302b478af Fix streaming in chat completions 2025-03-12 17:17:07 -07:00
Rohan Mehta 18b10f1371
Merge pull request #105 from openai/pr105 
Add max_tokens and documentation to model settings
 2025-03-12 13:45:21 -07:00
Rohan Mehta fde49cfcf7 Add max_tokens and documentation to model settings 2025-03-12 13:35:10 -07:00
Rohan Mehta f461bea2ce
Merge pull request #89 from eltociear/patch-1 
chore: update guardrail.py
 2025-03-12 13:30:56 -07:00
Rohan Mehta f2617595c6 Remove duplicated code 2025-03-12 13:08:00 -07:00
Muhammad Junaid 51d79bf141 fix: support assistant role in message conversion 
- The _Converter.items_to_messages method was incorrectly rejecting 'assistant'
as a valid role in conversation messages, causing runtime errors when processing
standard chat completion message formats.
- This fix enables proper handling of
complete conversation contexts that include both user and assistant messages.
 2025-03-12 21:10:03 +05:00
Ikko Eltociear Ashimine 77e359c0c7
chore: update guardrail.py 
minor fix
 2025-03-12 22:45:17 +09:00
Jai0401 0c33a24d8f fix: resolve linting issues 2025-03-12 16:00:20 +05:30
Jaimin Godhani a81da6788d
Update src/agents/tool.py 
Co-authored-by: Adrian Cole <64215+codefromthecrypt@users.noreply.github.com>
 2025-03-12 14:56:19 +05:30
Jai0401 b7e7fdee55 feat: Add strict mode option to function_schema and function_tool 2025-03-12 11:45:32 +05:30
Rohan Mehta be1c9495d6
Merge pull request #52 from sunrabbit123/patch-1 
Fix: remove unreached code in tool.py
 2025-03-11 22:09:07 -07:00
Rohan Mehta 90010a7eb5
Merge pull request #55 from the-praxs/fix/traces-spans-data 
fix: count `spans` and `traces` correctly
 2025-03-11 22:05:46 -07:00
Pratyush Shukla e5917724dc
resort to show total number of spans and traces 2025-03-12 10:20:15 +05:30
Pratyush Shukla 6962410976
Revert "count both spans and traces" 
This reverts commit d49bd89061.
 2025-03-12 10:19:14 +05:30
Pratyush Shukla d49bd89061
count both spans and traces 2025-03-12 10:14:01 +05:30
오병진 51a5be0212
Update tool.py 2025-03-12 13:13:48 +09:00
hironow 9b118a10c6
Merge branch 'main' into hironow/chore-doc-type 
# Conflicts:
#	tests/docs/config.md
#	tests/docs/guardrails.md
#	tests/docs/tracing.md
 2025-03-12 10:13:25 +09:00
hironow c827ecb788
Merge branch 'main' into hironow/chore-doc-type 2025-03-12 10:11:38 +09:00
Rohan Mehta c27d7b7d24 pin to openai 1.66.2, update tests 2025-03-11 15:26:06 -07:00
Rohan Mehta ce66d908a9 make format 2025-03-11 14:54:10 -07:00
dmitry-openai 9946b8b58d Merge pull request #8 from openai/repo/gitignore 
removing ds store, adding gitignore
 2025-03-11 11:17:37 -07:00
hironow 5a33e9b168
chore: fix type in doc and comment 2025-03-12 02:46:36 +09:00
Rohan Mehta aaec57a426 Initial commit 2025-03-11 09:42:28 -07:00