bridgekit 0.3.6__tar.gz → 0.3.8__tar.gz

{bridgekit-0.3.6 → bridgekit-0.3.8}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: bridgekit
-Version: 0.3.6
+Version: 0.3.8
 Summary: AI tools that make you a better data scientist, not a redundant one.
 License: MIT
 Project-URL: Homepage, https://usebridgekit.com
@@ -136,6 +136,9 @@ onboarding users to reporting as a growth lever.
 """
 
 print(evaluate(text))
+
+# Override for longer analyses
+print(evaluate(text, max_tokens=2048))
 ```
 
 **Output:**
@@ -190,6 +193,9 @@ Supports `.txt`, `.md`, `.pdf`, `.docx`, `.pptx`, and `.ipynb` files.
 from bridgekit import ask
 
 print(ask("what drove churn in Q3?", source="reports/"))
+
+# Override for longer responses
+print(ask("what drove churn in Q3?", source="reports/", max_tokens=2048))
 ```
 
 **From raw text:**
@@ -234,7 +240,7 @@ print(plan(
 ))
 ```
 
-`data_description` and `goal` are optional — the more context you provide, the more tailored the recommendation.
+`data_description`, `goal`, and `max_tokens` are optional — the more context you provide, the more tailored the recommendation.
 
 **`goal` examples:** `"causal inference"`, `"prediction"`, `"segmentation"`, `"hypothesis testing"`, `"exploration"`
 
@@ -297,6 +303,9 @@ print(redteam(text))
 # Or specify a stakeholder
 print(redteam(text, stakeholder="VP of Engineering"))
 print(redteam(text, stakeholder="VP of Marketing"))
+
+# Override for longer responses
+print(redteam(text, max_tokens=2048))
 ```
 
 Same writeup, different attack angles:
@@ -414,15 +423,39 @@ Bridgekit automatically detects the provider from model names:
 - Models starting with "gemini" → Google Gemini
 
 **Default models by provider:**
-- Anthropic: `claude-3-5-sonnet-20241022`
+- Anthropic: `claude-opus-4-8`
 - OpenAI: `gpt-4o`
 - Gemini: `gemini-1.5-pro`
 
 All tools support the same `provider` and `model` parameters:
-- `evaluate(text, provider=None, model=None)`
-- `plan(question, provider=None, model=None, ...)`
-- `ask(question, provider=None, model=None, ...)`
-- `redteam(text, provider=None, model=None, ...)`
+- `evaluate(text, provider=None, model=None, system_prompt=None)`
+- `plan(question, provider=None, model=None, ..., system_prompt=None)`
+- `ask(question, provider=None, model=None, ..., system_prompt=None)`
+- `redteam(text, provider=None, model=None, ..., system_prompt=None)`
+
+---
+
+## Custom System Prompts
+
+Every tool accepts an optional `system_prompt` parameter to override the default persona. Use this to adapt the tone or focus to a specific domain without changing anything else.
+
+```python
+from bridgekit import evaluate, plan, ask, redteam
+
+# Narrow the reviewer to a specific domain
+print(evaluate("my analysis", system_prompt="You are a skeptical PhD statistician focused only on methodology"))
+
+# Tailor the planner to a specific industry
+print(plan("my question", system_prompt="You are a data scientist specializing in healthcare analytics"))
+
+# Replace the red team persona entirely
+print(redteam("my analysis", system_prompt="You are a hostile regulator looking for compliance violations"))
+
+# Change the answering style for ask
+print(ask("my question", text="...", system_prompt="You are a financial analyst. Answer only in terms of revenue impact."))
+```
+
+When `system_prompt` is not provided, each tool uses its built-in default — existing behavior is unchanged.
 
 ---
 

{bridgekit-0.3.6 → bridgekit-0.3.8}/README.md

@@ -104,6 +104,9 @@ onboarding users to reporting as a growth lever.
 """
 
 print(evaluate(text))
+
+# Override for longer analyses
+print(evaluate(text, max_tokens=2048))
 ```
 
 **Output:**
@@ -158,6 +161,9 @@ Supports `.txt`, `.md`, `.pdf`, `.docx`, `.pptx`, and `.ipynb` files.
 from bridgekit import ask
 
 print(ask("what drove churn in Q3?", source="reports/"))
+
+# Override for longer responses
+print(ask("what drove churn in Q3?", source="reports/", max_tokens=2048))
 ```
 
 **From raw text:**
@@ -202,7 +208,7 @@ print(plan(
 ))
 ```
 
-`data_description` and `goal` are optional — the more context you provide, the more tailored the recommendation.
+`data_description`, `goal`, and `max_tokens` are optional — the more context you provide, the more tailored the recommendation.
 
 **`goal` examples:** `"causal inference"`, `"prediction"`, `"segmentation"`, `"hypothesis testing"`, `"exploration"`
 
@@ -265,6 +271,9 @@ print(redteam(text))
 # Or specify a stakeholder
 print(redteam(text, stakeholder="VP of Engineering"))
 print(redteam(text, stakeholder="VP of Marketing"))
+
+# Override for longer responses
+print(redteam(text, max_tokens=2048))
 ```
 
 Same writeup, different attack angles:
@@ -382,15 +391,39 @@ Bridgekit automatically detects the provider from model names:
 - Models starting with "gemini" → Google Gemini
 
 **Default models by provider:**
-- Anthropic: `claude-3-5-sonnet-20241022`
+- Anthropic: `claude-opus-4-8`
 - OpenAI: `gpt-4o`
 - Gemini: `gemini-1.5-pro`
 
 All tools support the same `provider` and `model` parameters:
-- `evaluate(text, provider=None, model=None)`
-- `plan(question, provider=None, model=None, ...)`
-- `ask(question, provider=None, model=None, ...)`
-- `redteam(text, provider=None, model=None, ...)`
+- `evaluate(text, provider=None, model=None, system_prompt=None)`
+- `plan(question, provider=None, model=None, ..., system_prompt=None)`
+- `ask(question, provider=None, model=None, ..., system_prompt=None)`
+- `redteam(text, provider=None, model=None, ..., system_prompt=None)`
+
+---
+
+## Custom System Prompts
+
+Every tool accepts an optional `system_prompt` parameter to override the default persona. Use this to adapt the tone or focus to a specific domain without changing anything else.
+
+```python
+from bridgekit import evaluate, plan, ask, redteam
+
+# Narrow the reviewer to a specific domain
+print(evaluate("my analysis", system_prompt="You are a skeptical PhD statistician focused only on methodology"))
+
+# Tailor the planner to a specific industry
+print(plan("my question", system_prompt="You are a data scientist specializing in healthcare analytics"))
+
+# Replace the red team persona entirely
+print(redteam("my analysis", system_prompt="You are a hostile regulator looking for compliance violations"))
+
+# Change the answering style for ask
+print(ask("my question", text="...", system_prompt="You are a financial analyst. Answer only in terms of revenue impact."))
+```
+
+When `system_prompt` is not provided, each tool uses its built-in default — existing behavior is unchanged.
 
 ---
 

{bridgekit-0.3.6 → bridgekit-0.3.8}/bridgekit/__init__.py

@@ -3,5 +3,5 @@ from .search import ask
 from .planner import plan
 from .redteam import redteam
 
-__version__ = "0.3.4"
+__version__ = "0.3.8"
 __all__ = ["evaluate", "ask", "plan", "redteam"]

bridgekit-0.3.8/bridgekit/cli.py

@@ -0,0 +1,105 @@
+import argparse
+import sys
+
+from .planner import plan
+from .reviewer import evaluate
+from .redteam import redteam
+from .search import ask
+
+
+def _add_provider_args(parser: argparse.ArgumentParser) -> None:
+    parser.add_argument("--provider", help='AI provider: "anthropic", "openai", or "gemini"')
+    parser.add_argument("--model", help="Specific model to use (e.g. claude-opus-4-8, gpt-4o)")
+
+
+def _cmd_plan(args: argparse.Namespace) -> None:
+    result = plan(
+        question=args.question,
+        data_description=args.data,
+        goal=args.goal,
+        provider=args.provider,
+        model=args.model,
+    )
+    print(result)
+
+
+def _cmd_review(args: argparse.Namespace) -> None:
+    result = evaluate(
+        text=args.text,
+        provider=args.provider,
+        model=args.model,
+    )
+    print(result)
+
+
+def _cmd_redteam(args: argparse.Namespace) -> None:
+    result = redteam(
+        text=args.text,
+        stakeholder=args.stakeholder,
+        provider=args.provider,
+        model=args.model,
+    )
+    print(result)
+
+
+def _cmd_search(args: argparse.Namespace) -> None:
+    if not args.source and not args.text:
+        print("error: provide --source or --text", file=sys.stderr)
+        sys.exit(1)
+    result = ask(
+        question=args.question,
+        source=args.source,
+        text=args.text,
+        provider=args.provider,
+        model=args.model,
+    )
+    print(result)
+
+
+def main() -> None:
+    parser = argparse.ArgumentParser(
+        prog="bridgekit",
+        description="AI tools for data scientists",
+    )
+    sub = parser.add_subparsers(dest="command", metavar="COMMAND")
+    sub.required = True
+
+    # plan
+    p_plan = sub.add_parser("plan", help="Recommend the right analytical approach")
+    p_plan.add_argument("question", help="The analytical question you want to answer")
+    p_plan.add_argument("--data", metavar="DESCRIPTION", help="Description of your available data")
+    p_plan.add_argument("--goal", help='Goal of the analysis (e.g. "prediction", "hypothesis testing")')
+    _add_provider_args(p_plan)
+    p_plan.set_defaults(func=_cmd_plan)
+
+    # review
+    p_review = sub.add_parser("review", help="Evaluate a data science analysis writeup")
+    p_review.add_argument("text", help="The analysis text to review")
+    _add_provider_args(p_review)
+    p_review.set_defaults(func=_cmd_review)
+
+    # redteam
+    p_redteam = sub.add_parser("redteam", help="Red-team an analysis from a skeptical stakeholder")
+    p_redteam.add_argument("text", help="The analysis text to red-team")
+    p_redteam.add_argument("--stakeholder", help='Stakeholder role (e.g. "VP of Finance")')
+    _add_provider_args(p_redteam)
+    p_redteam.set_defaults(func=_cmd_redteam)
+
+    # search
+    p_search = sub.add_parser("search", help="Ask a question across documents or text")
+    p_search.add_argument("question", help="The question to answer")
+    p_search.add_argument("--source", metavar="PATH", help="Folder of documents to search")
+    p_search.add_argument("--text", help="Raw text to search instead of a folder")
+    _add_provider_args(p_search)
+    p_search.set_defaults(func=_cmd_search)
+
+    args = parser.parse_args()
+    try:
+        args.func(args)
+    except (ValueError, EnvironmentError) as e:
+        print(f"error: {e}", file=sys.stderr)
+        sys.exit(1)
+
+
+if __name__ == "__main__":
+    main()

{bridgekit-0.3.6 → bridgekit-0.3.8}/bridgekit/config.py

@@ -11,7 +11,7 @@ class Provider(Enum):
 
 # Default models for each provider
 DEFAULT_MODELS = {
-    Provider.ANTHROPIC: "claude-opus-4-6",
+    Provider.ANTHROPIC: "claude-opus-4-8",
     Provider.OPENAI: "gpt-4o",
     Provider.GEMINI: "gemini-1.5-pro"
 }

{bridgekit-0.3.6 → bridgekit-0.3.8}/bridgekit/planner.py

@@ -29,7 +29,7 @@ ALTERNATIVES
 """
 
 
-def plan(question: str, data_description: str = None, goal: str = None, provider: str = None, model: str = None) -> str:
+def plan(question: str, data_description: str = None, goal: str = None, provider: str = None, model: str = None, system_prompt: str = None, max_tokens: int = 1024) -> str:
     """
     Recommend the right analytical approach for your problem.
 
@@ -41,6 +41,8 @@ def plan(question: str, data_description: str = None, goal: str = None, provider
         provider:         Optional. The AI provider to use ("anthropic", "openai", "gemini").
                           If not specified, defaults to "anthropic" or infers from model.
         model:            Optional. The specific model to use. If not specified, uses the provider's default.
+        system_prompt:    Optional. A custom system prompt to override the default planner persona.
+        max_tokens:       Optional. Maximum tokens in the response. Defaults to 1024.
 
     Returns:
         A structured analytical plan covering the recommended approach, assumptions,
@@ -62,8 +64,8 @@ def plan(question: str, data_description: str = None, goal: str = None, provider
 
     return create_message(
         provider=provider_enum,
-        system_prompt=SYSTEM_PROMPT,
+        system_prompt=system_prompt or SYSTEM_PROMPT,
         user_message=user_message,
         model=model,
-        max_tokens=1024
+        max_tokens=max_tokens
     )

{bridgekit-0.3.6 → bridgekit-0.3.8}/bridgekit/redteam.py

@@ -39,18 +39,21 @@ HARDEST QUESTION TO ANSWER
 """
 
 
-def redteam(text: str, stakeholder: str = None, provider: str = None, model: str = None) -> str:
+def redteam(text: str, stakeholder: str = None, provider: str = None, model: str = None, system_prompt: str = None, max_tokens: int = 1024) -> str:
     """
     Red-team a data science analysis writeup from the perspective of a skeptical stakeholder.
 
     Args:
-        text:        Your analysis writeup as a plain string.
-        stakeholder: Optional. The skeptical stakeholder role (e.g. "VP of Finance",
-                     "skeptical board member", "Chief Revenue Officer").
-                     Defaults to a generic skeptical senior executive.
-        provider:    Optional. The AI provider to use ("anthropic", "openai", "gemini").
-                     If not specified, defaults to "anthropic" or infers from model.
-        model:       Optional. The specific model to use. If not specified, uses the provider's default.
+        text:          Your analysis writeup as a plain string.
+        stakeholder:   Optional. The skeptical stakeholder role (e.g. "VP of Finance",
+                       "skeptical board member", "Chief Revenue Officer").
+                       Defaults to a generic skeptical senior executive.
+        provider:      Optional. The AI provider to use ("anthropic", "openai", "gemini").
+                       If not specified, defaults to "anthropic" or infers from model.
+        model:         Optional. The specific model to use. If not specified, uses the provider's default.
+        system_prompt: Optional. A custom system prompt to fully override the default red team persona.
+                       When provided, the stakeholder parameter is ignored.
+        max_tokens:    Optional. Maximum tokens in the response. Defaults to 1024.
 
     Returns:
         The 3-5 hardest critiques the stakeholder would make, plus the single
@@ -64,20 +67,20 @@ def redteam(text: str, stakeholder: str = None, provider: str = None, model: str
     if model is None:
         model = get_default_model(provider_enum)
 
-    stakeholder_label = stakeholder if stakeholder else "Skeptical Senior Executive"
-    stakeholder_desc = stakeholder if stakeholder else DEFAULT_STAKEHOLDER
-
-    system_prompt = SYSTEM_PROMPT_TEMPLATE.format(
-        stakeholder=stakeholder_desc,
-        stakeholder_label=stakeholder_label
-    )
+    if system_prompt is None:
+        stakeholder_label = stakeholder if stakeholder else "Skeptical Senior Executive"
+        stakeholder_desc = stakeholder if stakeholder else DEFAULT_STAKEHOLDER
+        system_prompt = SYSTEM_PROMPT_TEMPLATE.format(
+            stakeholder=stakeholder_desc,
+            stakeholder_label=stakeholder_label
+        )
 
     user_message = f"Red-team this analysis writeup:\n\n{text}"
-    
+
     return create_message(
         provider=provider_enum,
         system_prompt=system_prompt,
         user_message=user_message,
         model=model,
-        max_tokens=1024
+        max_tokens=max_tokens
     )

{bridgekit-0.3.6 → bridgekit-0.3.8}/bridgekit/reviewer.py

@@ -42,7 +42,7 @@ BOTTOM LINE
 [one sentence]
 """
 
-def evaluate(text: str, provider: str = None, model: str = None) -> str:
+def evaluate(text: str, provider: str = None, model: str = None, system_prompt: str = None, max_tokens: int = 1024) -> str:
     """
     Evaluate a data science analysis writeup and return structured feedback.
 
@@ -51,6 +51,8 @@ def evaluate(text: str, provider: str = None, model: str = None) -> str:
         provider: Optional. The AI provider to use ("anthropic", "openai", "gemini").
                  If not specified, defaults to "anthropic" or infers from model.
         model: Optional. The specific model to use. If not specified, uses the provider's default.
+        system_prompt: Optional. A custom system prompt to override the default reviewer persona.
+        max_tokens: Optional. Maximum tokens in the response. Defaults to 1024.
 
     Returns:
         Structured feedback across four dimensions.
@@ -64,11 +66,11 @@ def evaluate(text: str, provider: str = None, model: str = None) -> str:
         model = get_default_model(provider_enum)
 
     user_message = f"Please review this analysis writeup:\n\n{text}"
-    
+
     return create_message(
         provider=provider_enum,
-        system_prompt=SYSTEM_PROMPT,
+        system_prompt=system_prompt or SYSTEM_PROMPT,
         user_message=user_message,
         model=model,
-        max_tokens=1024
+        max_tokens=max_tokens
     )

{bridgekit-0.3.6 → bridgekit-0.3.8}/bridgekit/search.py

@@ -49,17 +49,26 @@ def _chunk(text: str) -> list[str]:
     return [c for c in chunks if c.strip()]
 
 
-def ask(question: str, source: str = None, text: str = None, provider: str = None, model: str = None) -> str:
+DEFAULT_SYSTEM_PROMPT = (
+    "You are a senior data scientist answering questions based on analysis reports. "
+    "Answer only from the provided context. Be specific and cite findings where relevant. "
+    "If the context does not contain enough information to answer, say so clearly."
+)
+
+
+def ask(question: str, source: str = None, text: str = None, provider: str = None, model: str = None, system_prompt: str = None, max_tokens: int = 1024) -> str:
     """
     Ask a question across a collection of analysis documents or raw text.
 
     Args:
-        question: The question to answer.
-        source:   Path to a folder containing .txt, .md, .pdf, .docx, .pptx, or .ipynb files.
-        text:     A raw text string to search instead of a folder.
-        provider: Optional. The AI provider to use ("anthropic", "openai", "gemini").
-                  If not specified, defaults to "anthropic" or infers from model.
-        model:    Optional. The specific model to use. If not specified, uses the provider's default.
+        question:      The question to answer.
+        source:        Path to a folder containing .txt, .md, .pdf, .docx, .pptx, or .ipynb files.
+        text:          A raw text string to search instead of a folder.
+        provider:      Optional. The AI provider to use ("anthropic", "openai", "gemini").
+                       If not specified, defaults to "anthropic" or infers from model.
+        model:         Optional. The specific model to use. If not specified, uses the provider's default.
+        system_prompt: Optional. A custom system prompt to override the default answering persona.
+        max_tokens:    Optional. Maximum tokens in the response. Defaults to 1024.
 
     Returns:
         An answer grounded in the provided documents.
@@ -107,15 +116,11 @@ def ask(question: str, source: str = None, text: str = None, provider: str = Non
 
     # Generate answer with specified provider
     user_message = f"Context from analysis reports:\n\n{context}\n\nQuestion: {question}"
-    
+
     return create_message(
         provider=provider_enum,
-        system_prompt=(
-            "You are a senior data scientist answering questions based on analysis reports. "
-            "Answer only from the provided context. Be specific and cite findings where relevant. "
-            "If the context does not contain enough information to answer, say so clearly."
-        ),
+        system_prompt=system_prompt or DEFAULT_SYSTEM_PROMPT,
         user_message=user_message,
         model=model,
-        max_tokens=1024
+        max_tokens=max_tokens
     )

{bridgekit-0.3.6 → bridgekit-0.3.8}/bridgekit.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: bridgekit
-Version: 0.3.6
+Version: 0.3.8
 Summary: AI tools that make you a better data scientist, not a redundant one.
 License: MIT
 Project-URL: Homepage, https://usebridgekit.com
@@ -136,6 +136,9 @@ onboarding users to reporting as a growth lever.
 """
 
 print(evaluate(text))
+
+# Override for longer analyses
+print(evaluate(text, max_tokens=2048))
 ```
 
 **Output:**
@@ -190,6 +193,9 @@ Supports `.txt`, `.md`, `.pdf`, `.docx`, `.pptx`, and `.ipynb` files.
 from bridgekit import ask
 
 print(ask("what drove churn in Q3?", source="reports/"))
+
+# Override for longer responses
+print(ask("what drove churn in Q3?", source="reports/", max_tokens=2048))
 ```
 
 **From raw text:**
@@ -234,7 +240,7 @@ print(plan(
 ))
 ```
 
-`data_description` and `goal` are optional — the more context you provide, the more tailored the recommendation.
+`data_description`, `goal`, and `max_tokens` are optional — the more context you provide, the more tailored the recommendation.
 
 **`goal` examples:** `"causal inference"`, `"prediction"`, `"segmentation"`, `"hypothesis testing"`, `"exploration"`
 
@@ -297,6 +303,9 @@ print(redteam(text))
 # Or specify a stakeholder
 print(redteam(text, stakeholder="VP of Engineering"))
 print(redteam(text, stakeholder="VP of Marketing"))
+
+# Override for longer responses
+print(redteam(text, max_tokens=2048))
 ```
 
 Same writeup, different attack angles:
@@ -414,15 +423,39 @@ Bridgekit automatically detects the provider from model names:
 - Models starting with "gemini" → Google Gemini
 
 **Default models by provider:**
-- Anthropic: `claude-3-5-sonnet-20241022`
+- Anthropic: `claude-opus-4-8`
 - OpenAI: `gpt-4o`
 - Gemini: `gemini-1.5-pro`
 
 All tools support the same `provider` and `model` parameters:
-- `evaluate(text, provider=None, model=None)`
-- `plan(question, provider=None, model=None, ...)`
-- `ask(question, provider=None, model=None, ...)`
-- `redteam(text, provider=None, model=None, ...)`
+- `evaluate(text, provider=None, model=None, system_prompt=None)`
+- `plan(question, provider=None, model=None, ..., system_prompt=None)`
+- `ask(question, provider=None, model=None, ..., system_prompt=None)`
+- `redteam(text, provider=None, model=None, ..., system_prompt=None)`
+
+---
+
+## Custom System Prompts
+
+Every tool accepts an optional `system_prompt` parameter to override the default persona. Use this to adapt the tone or focus to a specific domain without changing anything else.
+
+```python
+from bridgekit import evaluate, plan, ask, redteam
+
+# Narrow the reviewer to a specific domain
+print(evaluate("my analysis", system_prompt="You are a skeptical PhD statistician focused only on methodology"))
+
+# Tailor the planner to a specific industry
+print(plan("my question", system_prompt="You are a data scientist specializing in healthcare analytics"))
+
+# Replace the red team persona entirely
+print(redteam("my analysis", system_prompt="You are a hostile regulator looking for compliance violations"))
+
+# Change the answering style for ask
+print(ask("my question", text="...", system_prompt="You are a financial analyst. Answer only in terms of revenue impact."))
+```
+
+When `system_prompt` is not provided, each tool uses its built-in default — existing behavior is unchanged.
 
 ---
 

{bridgekit-0.3.6 → bridgekit-0.3.8}/bridgekit.egg-info/SOURCES.txt

@@ -2,6 +2,7 @@ LICENSE
 README.md
 pyproject.toml
 bridgekit/__init__.py
+bridgekit/cli.py
 bridgekit/config.py
 bridgekit/planner.py
 bridgekit/providers.py
@@ -11,10 +12,13 @@ bridgekit/search.py
 bridgekit.egg-info/PKG-INFO
 bridgekit.egg-info/SOURCES.txt
 bridgekit.egg-info/dependency_links.txt
+bridgekit.egg-info/entry_points.txt
 bridgekit.egg-info/requires.txt
 bridgekit.egg-info/top_level.txt
+tests/test_cli.py
 tests/test_config.py
 tests/test_planner.py
 tests/test_providers.py
+tests/test_redteam.py
 tests/test_reviewer.py
 tests/test_search.py

bridgekit-0.3.8/bridgekit.egg-info/entry_points.txt

@@ -0,0 +1,2 @@
+[console_scripts]
+bridgekit = bridgekit.cli:main

{bridgekit-0.3.6 → bridgekit-0.3.8}/pyproject.toml

@@ -7,7 +7,7 @@ include = ["bridgekit*"]
 
 [project]
 name = "bridgekit"
-version = "0.3.6"
+version = "0.3.8"
 description = "AI tools that make you a better data scientist, not a redundant one."
 readme = "README.md"
 requires-python = ">=3.9"
@@ -38,6 +38,9 @@ dev = [
     "pytest-mock>=3.0.0",
 ]
 
+[project.scripts]
+bridgekit = "bridgekit.cli:main"
+
 [project.urls]
 Homepage = "https://usebridgekit.com"
 Issues = "https://github.com/getbridgekit/bridgekit/issues"

bridgekit-0.3.8/tests/test_cli.py

@@ -0,0 +1,175 @@
+import sys
+import pytest
+from unittest.mock import patch
+
+
+FAKE_PLAN = "BRIDGEKIT ANALYSIS PLAN\n─────\nRECOMMENDED APPROACH\nUse a t-test."
+FAKE_REVIEW = "BRIDGEKIT ANALYSIS REVIEW\n─────\n1. CLARITY\n✅ STRONG Clear writing."
+FAKE_REDTEAM = "BRIDGEKIT RED TEAM\n─────\nCRITIQUE 1: Sample Size\nHARDEST QUESTION TO ANSWER\nWhat is n?"
+FAKE_SEARCH = "Based on the documents, the answer is 42."
+
+
+class TestPlanCommand:
+    def test_basic_question(self, capsys):
+        with patch("bridgekit.cli.plan", return_value=FAKE_PLAN) as mock_plan:
+            with patch("sys.argv", ["bridgekit", "plan", "should I use a t-test?"]):
+                from bridgekit.cli import main
+                main()
+        mock_plan.assert_called_once_with(
+            question="should I use a t-test?",
+            data_description=None,
+            goal=None,
+            provider=None,
+            model=None,
+        )
+        assert FAKE_PLAN in capsys.readouterr().out
+
+    def test_with_data_and_goal(self, capsys):
+        with patch("bridgekit.cli.plan", return_value=FAKE_PLAN) as mock_plan:
+            with patch("sys.argv", ["bridgekit", "plan", "my question",
+                                    "--data", "50 rows", "--goal", "compare means"]):
+                from bridgekit.cli import main
+                main()
+        mock_plan.assert_called_once_with(
+            question="my question",
+            data_description="50 rows",
+            goal="compare means",
+            provider=None,
+            model=None,
+        )
+
+    def test_with_provider_and_model(self):
+        with patch("bridgekit.cli.plan", return_value=FAKE_PLAN) as mock_plan:
+            with patch("sys.argv", ["bridgekit", "plan", "my question",
+                                    "--provider", "openai", "--model", "gpt-4o"]):
+                from bridgekit.cli import main
+                main()
+        mock_plan.assert_called_once_with(
+            question="my question",
+            data_description=None,
+            goal=None,
+            provider="openai",
+            model="gpt-4o",
+        )
+
+    def test_missing_question_exits(self):
+        with patch("sys.argv", ["bridgekit", "plan"]):
+            from bridgekit.cli import main
+            with pytest.raises(SystemExit):
+                main()
+
+    def test_environment_error_exits(self, capsys):
+        with patch("bridgekit.cli.plan", side_effect=EnvironmentError("ANTHROPIC_API_KEY not found")):
+            with patch("sys.argv", ["bridgekit", "plan", "my question"]):
+                from bridgekit.cli import main
+                with pytest.raises(SystemExit) as exc:
+                    main()
+        assert exc.value.code == 1
+        assert "ANTHROPIC_API_KEY" in capsys.readouterr().err
+
+
+class TestReviewCommand:
+    def test_basic_text(self, capsys):
+        with patch("bridgekit.cli.evaluate", return_value=FAKE_REVIEW) as mock_evaluate:
+            with patch("sys.argv", ["bridgekit", "review", "my analysis text"]):
+                from bridgekit.cli import main
+                main()
+        mock_evaluate.assert_called_once_with(
+            text="my analysis text",
+            provider=None,
+            model=None,
+        )
+        assert FAKE_REVIEW in capsys.readouterr().out
+
+    def test_missing_text_exits(self):
+        with patch("sys.argv", ["bridgekit", "review"]):
+            from bridgekit.cli import main
+            with pytest.raises(SystemExit):
+                main()
+
+
+class TestRedteamCommand:
+    def test_basic_text(self, capsys):
+        with patch("bridgekit.cli.redteam", return_value=FAKE_REDTEAM) as mock_redteam:
+            with patch("sys.argv", ["bridgekit", "redteam", "my analysis text"]):
+                from bridgekit.cli import main
+                main()
+        mock_redteam.assert_called_once_with(
+            text="my analysis text",
+            stakeholder=None,
+            provider=None,
+            model=None,
+        )
+        assert FAKE_REDTEAM in capsys.readouterr().out
+
+    def test_with_stakeholder(self):
+        with patch("bridgekit.cli.redteam", return_value=FAKE_REDTEAM) as mock_redteam:
+            with patch("sys.argv", ["bridgekit", "redteam", "my analysis text",
+                                    "--stakeholder", "VP of Finance"]):
+                from bridgekit.cli import main
+                main()
+        mock_redteam.assert_called_once_with(
+            text="my analysis text",
+            stakeholder="VP of Finance",
+            provider=None,
+            model=None,
+        )
+
+    def test_missing_text_exits(self):
+        with patch("sys.argv", ["bridgekit", "redteam"]):
+            from bridgekit.cli import main
+            with pytest.raises(SystemExit):
+                main()
+
+
+class TestSearchCommand:
+    def test_with_source(self, capsys):
+        with patch("bridgekit.cli.ask", return_value=FAKE_SEARCH) as mock_ask:
+            with patch("sys.argv", ["bridgekit", "search", "my question",
+                                    "--source", "./my_docs"]):
+                from bridgekit.cli import main
+                main()
+        mock_ask.assert_called_once_with(
+            question="my question",
+            source="./my_docs",
+            text=None,
+            provider=None,
+            model=None,
+        )
+        assert FAKE_SEARCH in capsys.readouterr().out
+
+    def test_with_text(self):
+        with patch("bridgekit.cli.ask", return_value=FAKE_SEARCH) as mock_ask:
+            with patch("sys.argv", ["bridgekit", "search", "my question",
+                                    "--text", "some raw text"]):
+                from bridgekit.cli import main
+                main()
+        mock_ask.assert_called_once_with(
+            question="my question",
+            source=None,
+            text="some raw text",
+            provider=None,
+            model=None,
+        )
+
+    def test_missing_source_and_text_exits(self, capsys):
+        with patch("sys.argv", ["bridgekit", "search", "my question"]):
+            from bridgekit.cli import main
+            with pytest.raises(SystemExit) as exc:
+                main()
+        assert exc.value.code == 1
+        assert "error" in capsys.readouterr().err
+
+    def test_missing_question_exits(self):
+        with patch("sys.argv", ["bridgekit", "search"]):
+            from bridgekit.cli import main
+            with pytest.raises(SystemExit):
+                main()
+
+
+class TestNoCommand:
+    def test_no_subcommand_exits(self):
+        with patch("sys.argv", ["bridgekit"]):
+            from bridgekit.cli import main
+            with pytest.raises(SystemExit):
+                main()

{bridgekit-0.3.6 → bridgekit-0.3.8}/tests/test_planner.py

@@ -159,6 +159,20 @@ class TestPlanOptionalParameters:
 
         assert isinstance(result, str)
 
+    def test_custom_system_prompt_reaches_api(self):
+        custom_prompt = "You are a data scientist specializing in healthcare analytics."
+        with patch.dict(os.environ, {"ANTHROPIC_API_KEY": "test-key"}):
+            with patch("anthropic.Anthropic") as MockAnthropic:
+                mock_client = MagicMock()
+                mock_client.messages.create.return_value = _make_mock_message(FAKE_RESPONSE)
+                MockAnthropic.return_value = mock_client
+
+                from bridgekit.planner import plan
+                plan("Should I use a t-test or ANOVA?", system_prompt=custom_prompt)
+
+                call_kwargs = mock_client.messages.create.call_args
+                assert call_kwargs.kwargs.get("system") == custom_prompt
+
     def test_all_parameters_included_in_api_call(self):
         with patch.dict(os.environ, {"ANTHROPIC_API_KEY": "test-key"}):
             with patch("anthropic.Anthropic") as MockAnthropic:
@@ -178,3 +192,33 @@ class TestPlanOptionalParameters:
                 content = str(messages_arg)
                 assert "5,000 users split 50/50." in content
                 assert "causal inference" in content
+
+
+class TestPlanMaxTokens:
+    """plan() should pass max_tokens through to the API."""
+
+    def test_default_max_tokens_is_1024(self):
+        with patch.dict(os.environ, {"ANTHROPIC_API_KEY": "test-key"}):
+            with patch("anthropic.Anthropic") as MockAnthropic:
+                mock_client = MagicMock()
+                mock_client.messages.create.return_value = _make_mock_message(FAKE_RESPONSE)
+                MockAnthropic.return_value = mock_client
+
+                from bridgekit.planner import plan
+                plan("Does our new onboarding flow increase upgrade rates?")
+
+                call_kwargs = mock_client.messages.create.call_args
+                assert call_kwargs.kwargs.get("max_tokens") == 1024
+
+    def test_custom_max_tokens_reaches_api(self):
+        with patch.dict(os.environ, {"ANTHROPIC_API_KEY": "test-key"}):
+            with patch("anthropic.Anthropic") as MockAnthropic:
+                mock_client = MagicMock()
+                mock_client.messages.create.return_value = _make_mock_message(FAKE_RESPONSE)
+                MockAnthropic.return_value = mock_client
+
+                from bridgekit.planner import plan
+                plan("Does our new onboarding flow increase upgrade rates?", max_tokens=2048)
+
+                call_kwargs = mock_client.messages.create.call_args
+                assert call_kwargs.kwargs.get("max_tokens") == 2048

bridgekit-0.3.8/tests/test_redteam.py

@@ -0,0 +1,185 @@
+import os
+import pytest
+from unittest.mock import MagicMock, patch
+
+
+# ---------------------------------------------------------------------------
+# Helpers
+# ---------------------------------------------------------------------------
+
+def _make_mock_message(text: str):
+    content_block = MagicMock()
+    content_block.text = text
+    message = MagicMock()
+    message.content = [content_block]
+    return message
+
+
+FAKE_RESPONSE = (
+    "BRIDGEKIT RED TEAM\n"
+    "─────────────────────────────────────────\n"
+    "STAKEHOLDER: Skeptical Senior Executive\n\n"
+    "CRITIQUE 1: Sample Size\n"
+    '❯ "How many users was this actually tested on?"\n'
+    "WHY IT LANDS: No sample size is mentioned anywhere.\n"
+    "TO ADDRESS: Report n for each group with a power calculation.\n\n"
+    "CRITIQUE 2: Causation vs Correlation\n"
+    '❯ "You\'re assuming the feature caused this lift — prove it."\n'
+    "WHY IT LANDS: No control group is described.\n"
+    "TO ADDRESS: Show the experimental design with random assignment.\n\n"
+    "CRITIQUE 3: Business Impact\n"
+    '❯ "What does a 5% lift actually mean in dollars?"\n'
+    "WHY IT LANDS: Directional claims are not quantified.\n"
+    "TO ADDRESS: Translate the metric into revenue or cost terms.\n\n"
+    "─────────────────────────────────────────\n"
+    "HARDEST QUESTION TO ANSWER\n"
+    "What is the p-value and did you correct for multiple comparisons?"
+)
+
+
+# ---------------------------------------------------------------------------
+# Tests
+# ---------------------------------------------------------------------------
+
+class TestRedteamReturnsString:
+    """redteam() should return a non-empty string."""
+
+    def test_returns_string(self):
+        with patch.dict(os.environ, {"ANTHROPIC_API_KEY": "test-key"}):
+            with patch("anthropic.Anthropic") as MockAnthropic:
+                mock_client = MagicMock()
+                mock_client.messages.create.return_value = _make_mock_message(FAKE_RESPONSE)
+                MockAnthropic.return_value = mock_client
+
+                from bridgekit.redteam import redteam
+                result = redteam("We ran an A/B test and saw a 5% lift in conversions.")
+
+        assert isinstance(result, str)
+        assert len(result) > 0
+
+
+class TestRedteamOutputStructure:
+    """redteam() output should contain the required section headers."""
+
+    def test_output_contains_critique(self):
+        with patch.dict(os.environ, {"ANTHROPIC_API_KEY": "test-key"}):
+            with patch("anthropic.Anthropic") as MockAnthropic:
+                mock_client = MagicMock()
+                mock_client.messages.create.return_value = _make_mock_message(FAKE_RESPONSE)
+                MockAnthropic.return_value = mock_client
+
+                from bridgekit.redteam import redteam
+                result = redteam("We ran an A/B test and saw a 5% lift in conversions.")
+
+        assert "CRITIQUE" in result
+
+    def test_output_contains_hardest_question(self):
+        with patch.dict(os.environ, {"ANTHROPIC_API_KEY": "test-key"}):
+            with patch("anthropic.Anthropic") as MockAnthropic:
+                mock_client = MagicMock()
+                mock_client.messages.create.return_value = _make_mock_message(FAKE_RESPONSE)
+                MockAnthropic.return_value = mock_client
+
+                from bridgekit.redteam import redteam
+                result = redteam("We ran an A/B test and saw a 5% lift in conversions.")
+
+        assert "HARDEST QUESTION" in result
+
+
+class TestRedteamMissingApiKey:
+    """redteam() should raise EnvironmentError when ANTHROPIC_API_KEY is absent."""
+
+    def test_raises_environment_error_when_key_missing(self):
+        env = {k: v for k, v in os.environ.items() if k != "ANTHROPIC_API_KEY"}
+        with patch.dict(os.environ, env, clear=True):
+            from bridgekit.redteam import redteam
+            with pytest.raises(EnvironmentError):
+                redteam("Some analysis text.")
+
+    def test_error_message_mentions_key(self):
+        env = {k: v for k, v in os.environ.items() if k != "ANTHROPIC_API_KEY"}
+        with patch.dict(os.environ, env, clear=True):
+            from bridgekit.redteam import redteam
+            with pytest.raises(EnvironmentError, match="ANTHROPIC_API_KEY"):
+                redteam("Some analysis text.")
+
+
+class TestRedteamEmptyInput:
+    """redteam() should raise ValueError for empty or whitespace-only input."""
+
+    def test_empty_string_raises_value_error(self):
+        with patch.dict(os.environ, {"ANTHROPIC_API_KEY": "test-key"}):
+            from bridgekit.redteam import redteam
+            with pytest.raises(ValueError, match="empty"):
+                redteam("")
+
+    def test_whitespace_only_raises_value_error(self):
+        with patch.dict(os.environ, {"ANTHROPIC_API_KEY": "test-key"}):
+            from bridgekit.redteam import redteam
+            with pytest.raises(ValueError, match="empty"):
+                redteam("   ")
+
+
+class TestRedteamStakeholder:
+    """redteam() should include a custom stakeholder in the system prompt."""
+
+    def test_custom_stakeholder_reaches_system_prompt(self):
+        with patch.dict(os.environ, {"ANTHROPIC_API_KEY": "test-key"}):
+            with patch("anthropic.Anthropic") as MockAnthropic:
+                mock_client = MagicMock()
+                mock_client.messages.create.return_value = _make_mock_message(FAKE_RESPONSE)
+                MockAnthropic.return_value = mock_client
+
+                from bridgekit.redteam import redteam
+                redteam("Some analysis text.", stakeholder="VP of Finance")
+
+                call_kwargs = mock_client.messages.create.call_args
+                assert "VP of Finance" in call_kwargs.kwargs.get("system", "")
+
+
+class TestRedteamCustomSystemPrompt:
+    """redteam() should forward a custom system_prompt to the API, ignoring stakeholder."""
+
+    def test_custom_system_prompt_reaches_api(self):
+        custom_prompt = "You are a hostile regulator looking for compliance violations."
+        with patch.dict(os.environ, {"ANTHROPIC_API_KEY": "test-key"}):
+            with patch("anthropic.Anthropic") as MockAnthropic:
+                mock_client = MagicMock()
+                mock_client.messages.create.return_value = _make_mock_message(FAKE_RESPONSE)
+                MockAnthropic.return_value = mock_client
+
+                from bridgekit.redteam import redteam
+                redteam("Some analysis text.", system_prompt=custom_prompt)
+
+                call_kwargs = mock_client.messages.create.call_args
+                assert call_kwargs.kwargs.get("system") == custom_prompt
+
+
+class TestRedteamMaxTokens:
+    """redteam() should pass max_tokens through to the API."""
+
+    def test_default_max_tokens_is_1024(self):
+        with patch.dict(os.environ, {"ANTHROPIC_API_KEY": "test-key"}):
+            with patch("anthropic.Anthropic") as MockAnthropic:
+                mock_client = MagicMock()
+                mock_client.messages.create.return_value = _make_mock_message(FAKE_RESPONSE)
+                MockAnthropic.return_value = mock_client
+
+                from bridgekit.redteam import redteam
+                redteam("Some analysis text.")
+
+                call_kwargs = mock_client.messages.create.call_args
+                assert call_kwargs.kwargs.get("max_tokens") == 1024
+
+    def test_custom_max_tokens_reaches_api(self):
+        with patch.dict(os.environ, {"ANTHROPIC_API_KEY": "test-key"}):
+            with patch("anthropic.Anthropic") as MockAnthropic:
+                mock_client = MagicMock()
+                mock_client.messages.create.return_value = _make_mock_message(FAKE_RESPONSE)
+                MockAnthropic.return_value = mock_client
+
+                from bridgekit.redteam import redteam
+                redteam("Some analysis text.", max_tokens=2048)
+
+                call_kwargs = mock_client.messages.create.call_args
+                assert call_kwargs.kwargs.get("max_tokens") == 2048

{bridgekit-0.3.6 → bridgekit-0.3.8}/tests/test_reviewer.py

@@ -158,3 +158,51 @@ class TestEvaluateApiCallShape:
                 messages_arg = call_kwargs.kwargs.get("messages") or call_kwargs.args[0]
                 content = str(messages_arg)
                 assert user_text in content
+
+
+class TestEvaluateCustomSystemPrompt:
+    """evaluate() should forward a custom system_prompt to the API."""
+
+    def test_custom_system_prompt_reaches_api(self):
+        custom_prompt = "You are a skeptical PhD statistician focused only on methodology."
+        with patch.dict(os.environ, {"ANTHROPIC_API_KEY": "test-key"}):
+            with patch("anthropic.Anthropic") as MockAnthropic:
+                mock_client = MagicMock()
+                mock_client.messages.create.return_value = _make_mock_message(FAKE_RESPONSE)
+                MockAnthropic.return_value = mock_client
+
+                from bridgekit.reviewer import evaluate
+                evaluate("Some analysis text.", system_prompt=custom_prompt)
+
+                call_kwargs = mock_client.messages.create.call_args
+                assert call_kwargs.kwargs.get("system") == custom_prompt
+
+
+class TestEvaluateMaxTokens:
+    """evaluate() should pass max_tokens through to the API."""
+
+    def test_default_max_tokens_is_1024(self):
+        with patch.dict(os.environ, {"ANTHROPIC_API_KEY": "test-key"}):
+            with patch("anthropic.Anthropic") as MockAnthropic:
+                mock_client = MagicMock()
+                mock_client.messages.create.return_value = _make_mock_message(FAKE_RESPONSE)
+                MockAnthropic.return_value = mock_client
+
+                from bridgekit.reviewer import evaluate
+                evaluate("Some analysis text.")
+
+                call_kwargs = mock_client.messages.create.call_args
+                assert call_kwargs.kwargs.get("max_tokens") == 1024
+
+    def test_custom_max_tokens_reaches_api(self):
+        with patch.dict(os.environ, {"ANTHROPIC_API_KEY": "test-key"}):
+            with patch("anthropic.Anthropic") as MockAnthropic:
+                mock_client = MagicMock()
+                mock_client.messages.create.return_value = _make_mock_message(FAKE_RESPONSE)
+                MockAnthropic.return_value = mock_client
+
+                from bridgekit.reviewer import evaluate
+                evaluate("Some analysis text.", max_tokens=2048)
+
+                call_kwargs = mock_client.messages.create.call_args
+                assert call_kwargs.kwargs.get("max_tokens") == 2048

{bridgekit-0.3.6 → bridgekit-0.3.8}/tests/test_search.py

@@ -222,6 +222,27 @@ class TestAskWithSourceFolder:
 
                     assert mock_client.messages.create.call_count == 1
 
+    def test_custom_system_prompt_reaches_api(self):
+        custom_prompt = "You are a financial analyst. Answer only in terms of revenue impact."
+        mock_chromadb, mock_ef = _make_mock_chromadb()
+
+        with patch.dict(os.environ, {"ANTHROPIC_API_KEY": "test-key"}):
+            with patch("anthropic.Anthropic") as MockAnthropic, \
+                 patch("chromadb.Client", mock_chromadb.Client), \
+                 patch(
+                     "chromadb.utils.embedding_functions.SentenceTransformerEmbeddingFunction",
+                     mock_ef,
+                 ):
+                mock_client = MagicMock()
+                mock_client.messages.create.return_value = _make_mock_message(FAKE_ANSWER)
+                MockAnthropic.return_value = mock_client
+
+                from bridgekit.search import ask
+                ask("What was revenue?", text="Revenue was $5M.", system_prompt=custom_prompt)
+
+                call_kwargs = mock_client.messages.create.call_args
+                assert call_kwargs.kwargs.get("system") == custom_prompt
+
     def test_source_folder_empty_raises_value_error(self):
         with tempfile.TemporaryDirectory() as tmpdir:
             # Folder exists but has no supported files
@@ -236,3 +257,45 @@ class TestAskWithSourceFolder:
                     from bridgekit.search import ask
                     with pytest.raises(ValueError, match="No content found"):
                         ask("What happened?", source=tmpdir)
+
+
+class TestAskMaxTokens:
+    """ask() should pass max_tokens through to the API."""
+
+    def test_default_max_tokens_is_1024(self):
+        mock_chromadb, mock_ef = _make_mock_chromadb()
+        with patch.dict(os.environ, {"ANTHROPIC_API_KEY": "test-key"}):
+            with patch("anthropic.Anthropic") as MockAnthropic, \
+                 patch("chromadb.Client", mock_chromadb.Client), \
+                 patch(
+                     "chromadb.utils.embedding_functions.SentenceTransformerEmbeddingFunction",
+                     mock_ef,
+                 ):
+                mock_client = MagicMock()
+                mock_client.messages.create.return_value = _make_mock_message(FAKE_ANSWER)
+                MockAnthropic.return_value = mock_client
+
+                from bridgekit.search import ask
+                ask("What was the conversion rate?", text="The conversion rate increased by 12%.")
+
+                call_kwargs = mock_client.messages.create.call_args
+                assert call_kwargs.kwargs.get("max_tokens") == 1024
+
+    def test_custom_max_tokens_reaches_api(self):
+        mock_chromadb, mock_ef = _make_mock_chromadb()
+        with patch.dict(os.environ, {"ANTHROPIC_API_KEY": "test-key"}):
+            with patch("anthropic.Anthropic") as MockAnthropic, \
+                 patch("chromadb.Client", mock_chromadb.Client), \
+                 patch(
+                     "chromadb.utils.embedding_functions.SentenceTransformerEmbeddingFunction",
+                     mock_ef,
+                 ):
+                mock_client = MagicMock()
+                mock_client.messages.create.return_value = _make_mock_message(FAKE_ANSWER)
+                MockAnthropic.return_value = mock_client
+
+                from bridgekit.search import ask
+                ask("What was the conversion rate?", text="The conversion rate increased by 12%.", max_tokens=2048)
+
+                call_kwargs = mock_client.messages.create.call_args
+                assert call_kwargs.kwargs.get("max_tokens") == 2048