verifiers 0.1.13.dev5__tar.gz → 0.1.13.dev7__tar.gz

{verifiers-0.1.13.dev5 → verifiers-0.1.13.dev7}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: verifiers
-Version: 0.1.13.dev5
+Version: 0.1.13.dev7
 Summary: Verifiers: Environments for LLM Reinforcement Learning
 Project-URL: Homepage, https://github.com/primeintellect-ai/verifiers
 Project-URL: Documentation, https://github.com/primeintellect-ai/verifiers

{verifiers-0.1.13.dev5 → verifiers-0.1.13.dev7}/verifiers/__init__.py

@@ -1,4 +1,4 @@
-__version__ = "0.1.13.dev5"
+__version__ = "0.1.13.dev7"
 
 import importlib
 import os

{verifiers-0.1.13.dev5 → verifiers-0.1.13.dev7}/verifiers/envs/experimental/composable/harness.py

@@ -88,7 +88,7 @@ class Harness:
         ``ComposableEnv`` auto-registers a ``ToolMonitorRubric`` that
         counts calls to each named tool (plus a total) from the
         assistant messages the harness emits into the trajectory.
-        Example: ``["ipython", "summarize"]`` for the RLM harness.
+        Example: ``["ipython"]`` for the RLM harness.
     environment_vars:
         Harness-owned environment variables for the sandbox. Merged by
         ``ComposableEnv`` between the caller-supplied ``environment_vars=``

{verifiers-0.1.13.dev5 → verifiers-0.1.13.dev7}/verifiers/envs/experimental/composable/harnesses/__init__.py

@@ -1,4 +1,5 @@
 from verifiers.envs.experimental.composable.harnesses.rlm import (
+    DEFAULT_RLM_EXEC_TIMEOUT,
     DEFAULT_RLM_MAX_TURNS,
     DEFAULT_RLM_REF,
     DEFAULT_RLM_REPO_URL,
@@ -31,6 +32,7 @@ __all__ = [
     "DEFAULT_RLM_REF",
     "DEFAULT_RLM_REPO_URL",
     "DEFAULT_RLM_MAX_TURNS",
+    "DEFAULT_RLM_EXEC_TIMEOUT",
     "opencode_harness",
     "build_opencode_install_script",
     "build_opencode_config",

{verifiers-0.1.13.dev5 → verifiers-0.1.13.dev7}/verifiers/envs/experimental/composable/harnesses/rlm.py

@@ -15,6 +15,7 @@ from verifiers.envs.experimental.utils.git_checkout_cache import (
 DEFAULT_RLM_REPO_URL = "github.com/PrimeIntellect-ai/rlm.git"
 DEFAULT_RLM_REF = "main"
 DEFAULT_RLM_MAX_TURNS = 100
+DEFAULT_RLM_EXEC_TIMEOUT = 300
 DEFAULT_APPEND_TO_SYSTEM_PROMPT_PATH = "/task/append_to_system_prompt.txt"
 DEFAULT_RLM_CHECKOUT_PATH = "/tmp/rlm-checkout"
 DEFAULT_RLM_CHECKOUT_UPLOAD_NAME = "rlm_checkout"
@@ -98,6 +99,9 @@ def rlm_harness(
     instruction_path: str = "/task/instruction.md",
     rlm_repo_url: str = DEFAULT_RLM_REPO_URL,
     rlm_ref: str = DEFAULT_RLM_REF,
+    rlm_max_turns: int = DEFAULT_RLM_MAX_TURNS,
+    rlm_exec_timeout: int = DEFAULT_RLM_EXEC_TIMEOUT,
+    summarize_at_tokens: int | None = None,
     append_to_system_prompt: str | None = None,
     local_checkout: str | Path | None = None,
     gh_token: str | None = None,
@@ -106,13 +110,23 @@ def rlm_harness(
 ) -> Harness:
     """Build an RLM harness.
 
-    ``rlm_tools`` is the single source of truth for which builtin tools are
-    active. The same list drives both ``Harness.tool_names`` (so
-    ``ToolMonitorRubric`` tracks exactly the active tools) and
-    ``Harness.environment_vars["RLM_TOOLS"]`` (so the RLM sandbox advertises
-    the same set to the model). Callers do not need to — and should not —
-    add ``RLM_TOOLS`` to ``ComposableEnv(environment_vars=...)`` themselves;
-    the harness owns it.
+    The harness is the single source of truth for every ``RLM_*`` sandbox
+    env var the RLM subprocess reads. Kwargs map 1:1 onto env vars written
+    to ``Harness.environment_vars`` and merged into the sandbox by
+    ``ComposableEnv`` (harness-wins):
+
+    - ``rlm_tools`` → ``RLM_TOOLS`` (also drives ``Harness.tool_names`` so
+      ``ToolMonitorRubric`` tracks exactly the active tools)
+    - ``rlm_max_turns`` → ``RLM_MAX_TURNS``
+    - ``rlm_exec_timeout`` → ``RLM_EXEC_TIMEOUT``
+    - ``summarize_at_tokens`` → ``RLM_SUMMARIZE_AT_TOKENS``: when set to
+      a positive int, rlm auto-compacts the current branch once the
+      prompt_tokens of a turn reach the threshold. ``None`` disables
+      auto-compaction.
+
+    Callers do not need to — and should not — add these keys to
+    ``ComposableEnv(environment_vars=...)`` themselves; pass the kwargs
+    here and the harness owns the env var plumbing.
 
     ``allow_git`` defaults to False, mirroring opencode's bash tool. When
     False, a ``/usr/local/bin/git`` shim is uploaded that refuses on any
@@ -142,7 +156,7 @@ def rlm_harness(
         resolved_upload_dirs = upload_dirs
         return resolved_upload_dirs
 
-    tool_names = list(rlm_tools) if rlm_tools is not None else ["ipython", "summarize"]
+    tool_names = list(rlm_tools) if rlm_tools is not None else ["ipython"]
 
     post_install_uploads: dict[str, str] | None = None
     post_install_script: str | None = None
@@ -150,6 +164,15 @@ def rlm_harness(
         post_install_uploads = {"/usr/local/bin/git": _GIT_SHIM_BODY}
         post_install_script = "chmod +x /usr/local/bin/git"
 
+    environment_vars: dict[str, str] = {
+        "RLM_TOOLS": ",".join(tool_names),
+        "RLM_MAX_TURNS": str(rlm_max_turns),
+        "RLM_EXEC_TIMEOUT": str(rlm_exec_timeout),
+    }
+    summarize_env = _format_summarize_at_tokens(summarize_at_tokens)
+    if summarize_env is not None:
+        environment_vars["RLM_SUMMARIZE_AT_TOKENS"] = summarize_env
+
     return Harness(
         install_script=build_install_script(),
         run_command=build_run_command(instruction_path, workdir),
@@ -163,7 +186,26 @@ def rlm_harness(
         metrics_key="metrics",
         metrics_prefix="rlm_",
         tool_names=tool_names,
-        environment_vars={"RLM_TOOLS": ",".join(tool_names)},
+        environment_vars=environment_vars,
         post_install_uploads=post_install_uploads,
         post_install_script=post_install_script,
     )
+
+
+def _format_summarize_at_tokens(value: int | None) -> str | None:
+    """Format ``summarize_at_tokens`` as the ``RLM_SUMMARIZE_AT_TOKENS`` string.
+
+    Returns ``None`` when auto-compaction should be disabled (matches what
+    the engine expects when the env var is absent). Rejects bad shapes
+    here so configuration errors surface at harness-build time rather
+    than deep inside the sandbox.
+    """
+    if value is None:
+        return None
+    if isinstance(value, bool) or not isinstance(value, int):
+        raise ValueError(
+            f"summarize_at_tokens must be an int or None (got {type(value).__name__})"
+        )
+    if value <= 0:
+        raise ValueError(f"summarize_at_tokens must be positive (got {value})")
+    return str(value)