aru-code 0.25.0__tar.gz → 0.25.2__tar.gz

{aru_code-0.25.0/aru_code.egg-info → aru_code-0.25.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: aru-code
-Version: 0.25.0
+Version: 0.25.2
 Summary: A Claude Code clone built with Agno agents
 Author-email: Estevao <estevaofon@gmail.com>
 License-Expression: MIT

aru_code-0.25.2/aru/__init__.py

@@ -0,0 +1 @@
+__version__ = "0.25.2"

{aru_code-0.25.0 → aru_code-0.25.2}/aru/agent_factory.py

@@ -182,7 +182,7 @@ def create_general_agent(
         tools=tools,
         instructions=instructions,
         markdown=True,
-        tool_call_limit=20,
+        tool_call_limit=None,
     )
 
 
@@ -216,5 +216,5 @@ def create_custom_agent_instance(agent_def: CustomAgent, session: Session,
         tools=tools,
         instructions=instructions,
         markdown=True,
-        tool_call_limit=agent_def.max_turns or 20,
+        tool_call_limit=agent_def.max_turns,
     )

{aru_code-0.25.0 → aru_code-0.25.2}/aru/agents/base.py

@@ -139,13 +139,34 @@ split into subtasks grouped by concern (e.g. "Create model files", "Create route
 - Read files before editing them
 - Use edit_file for targeted changes (preferred over rewriting entire files)
 - Use write_file only for new files or complete rewrites
-- Run existing tests after changes when applicable
-- **When adding or modifying unit tests, ALWAYS run them to verify they pass before finishing.**
 - Keep changes minimal and focused on the task
 - Do not add unnecessary comments, docstrings, or refactoring beyond what was asked
 - **One ask = one deliverable.** If asked for one function, write one function. \
   Helper functions are NOT implicit — do not add them unless explicitly requested.
 
+## Verification — run it before claiming done
+
+Never mark a task done on faith. Prove the change works by running something that \
+exercises it — invoke the function, trigger the code path, fire the test, and read \
+what comes back. Editing a file is not the same as confirming the edit is correct. \
+When running something is genuinely impossible (no runnable harness, sandbox blocks \
+execution, external service unreachable), state that limitation plainly instead of \
+calling the work done.
+
+Concrete patterns:
+- **After a bug fix**: reproduce the failing case and confirm it now passes.
+- **After writing a plugin/tool/module**: invoke it inline with a realistic input and \
+  inspect the output — don't stop at "it imports". Use `bash` to run a one-shot probe, \
+  e.g. `python -c "from mod import fn; print(repr(fn(<realistic input>)))"`, read the \
+  `repr()`, and iterate until it matches expectations.
+- **After a refactor**: run the existing test suite — or if none exists, exercise the \
+  changed path manually and read the result.
+- **After adding or modifying unit tests**: ALWAYS run them before finishing.
+
+A good colleague doesn't stop at "it compiles" — they run it, read the output, and fix \
+the gap between what they wrote and what they meant. Each bug surfaced by a 10-second \
+inline probe is a bug the user never has to report.
+
 ## Reading strategy — read, edit, test
 
 1. **Need a specific pattern?** → `grep_search(pattern, file_glob="*.py")` — default 10 lines context. \
@@ -208,7 +229,28 @@ Every tool call accumulates its result in your context window. Use the minimum n
 **Stop early**: Once you have enough information to act, stop exploring and start working. \
 Batch what you need upfront, then execute.
 
-**When adding or modifying unit tests, ALWAYS run them to verify they pass before finishing.**
+## Verification — run it before claiming done
+
+Never mark a task done on faith. Prove the change works by running something that \
+exercises it — invoke the function, trigger the code path, fire the test, and read \
+what comes back. Editing a file is not the same as confirming the edit is correct. \
+When running something is genuinely impossible (no runnable harness, sandbox blocks \
+execution, external service unreachable), state that limitation plainly instead of \
+calling the work done.
+
+Concrete patterns:
+- **After a bug fix**: reproduce the failing case and confirm it now passes.
+- **After writing a plugin/tool/module**: invoke it inline with a realistic input and \
+  inspect the output — don't stop at "it imports". Use `bash` to run a one-shot probe, \
+  e.g. `python -c "from mod import fn; print(repr(fn(<realistic input>)))"`, read the \
+  `repr()`, and iterate until it matches expectations.
+- **After a refactor**: run the existing test suite — or if none exists, exercise the \
+  changed path manually and read the result.
+- **After adding or modifying unit tests**: ALWAYS run them before finishing.
+
+A good colleague doesn't stop at "it compiles" — they run it, read the output, and fix \
+the gap between what they wrote and what they meant. Each bug surfaced by a 10-second \
+inline probe is a bug the user never has to report.
 
 ## Delegation strategy — CRITICAL for context efficiency
 

{aru_code-0.25.0 → aru_code-0.25.2}/aru/agents/explorer.py

@@ -89,5 +89,5 @@ def create_explorer(task: str, context: str = "") -> Agent:
         tools=EXPLORER_TOOLS,
         instructions=instructions,
         markdown=True,
-        tool_call_limit=15,
+        tool_call_limit=None,
     )

{aru_code-0.25.0 → aru_code-0.25.2}/aru/agents/planner.py

@@ -85,5 +85,5 @@ def create_planner(model_ref: str = "anthropic/claude-sonnet-4-5", extra_instruc
             compress_tool_results=True,
             compress_tool_results_limit=25,
         ),
-        tool_call_limit=20,
+        tool_call_limit=None,
     )

{aru_code-0.25.0 → aru_code-0.25.2/aru_code.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: aru-code
-Version: 0.25.0
+Version: 0.25.2
 Summary: A Claude Code clone built with Agno agents
 Author-email: Estevao <estevaofon@gmail.com>
 License-Expression: MIT

{aru_code-0.25.0 → aru_code-0.25.2}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "aru-code"
-version = "0.25.0"
+version = "0.25.2"
 description = "A Claude Code clone built with Agno agents"
 readme = "README.md"
 license = "MIT"

aru_code-0.25.0/aru/__init__.py

@@ -1 +0,0 @@
-__version__ = "0.25.0"