Let uv-installed optimizer harnesses install as the unprivileged agent user

`uv tool install` symlinks a harness's entry point into /usr/local/bin, which the
optimizer user cannot write, so the install step dies with "Failed to install
executable ... Permission denied (os error 13)" and the trial ends in
NonZeroAgentExitCodeError before any evaluation runs. Pointing UV_TOOL_BIN_DIR at
the user's own bin directory fixes it.

This blocks both literature-standard harnesses -- mini-swe-agent and swe-agent are
each installed this way -- while claude-code and opencode are npm/nvm-based and
never hit it, so the failure only appears when the harness dimension of the grid
changes, and reads like "this harness does not work with vero" rather than a
container-permissions problem. Found by vero/examples/harness-conformance in 91
seconds; mini-swe-agent then scored 1.0 on the same check in 4m57s, with 25
producer requests all 200 on chat/completions and no denials.

Applied to all five benchmarks rather than the one about to be run, since the
first mini-swe-agent cell of the grid would otherwise hit it whichever benchmark
came first.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

Author: varunursekar

harness-engineering-bench/browsecomp-plus/baseline/build.yaml

@@ -135,6 +135,12 @@ agent_env:
   BASH_DEFAULT_TIMEOUT_MS: "14400000"  # same as max: an un-timed eval must still block
   ENABLE_BACKGROUND_TASKS: "0"
   FORCE_AUTO_BACKGROUND_TASKS: "0"
+  # Harnesses installed with `uv tool install` (mini-swe-agent, swe-agent)
+  # default to symlinking their entry point into /usr/local/bin, which the
+  # unprivileged optimizer user cannot write: "Failed to install executable
+  # ... Permission denied". npm/nvm-based harnesses (claude-code, opencode)
+  # are unaffected, so this only bites when the harness changes.
+  UV_TOOL_BIN_DIR: "/home/agent/.local/bin"
 
 task_environment:
   BROWSECOMP_JUDGE_MODEL: gpt-4.1

harness-engineering-bench/gaia/baseline/build.yaml

@@ -111,6 +111,12 @@ agent_env:
   BASH_DEFAULT_TIMEOUT_MS: "3600000"  # same as max: an un-timed eval must still block
   ENABLE_BACKGROUND_TASKS: "0"
   FORCE_AUTO_BACKGROUND_TASKS: "0"
+  # Harnesses installed with `uv tool install` (mini-swe-agent, swe-agent)
+  # default to symlinking their entry point into /usr/local/bin, which the
+  # unprivileged optimizer user cannot write: "Failed to install executable
+  # ... Permission denied". npm/nvm-based harnesses (claude-code, opencode)
+  # are unaffected, so this only bites when the harness changes.
+  UV_TOOL_BIN_DIR: "/home/agent/.local/bin"
 
 
 wandb:

harness-engineering-bench/officeqa/baseline/build.yaml

@@ -117,6 +117,12 @@ agent_env:
   BASH_DEFAULT_TIMEOUT_MS: "10800000"  # same as max: an un-timed eval must still block
   ENABLE_BACKGROUND_TASKS: "0"         # gates auto-backgrounding only (see above)
   FORCE_AUTO_BACKGROUND_TASKS: "0"
+  # Harnesses installed with `uv tool install` (mini-swe-agent, swe-agent)
+  # default to symlinking their entry point into /usr/local/bin, which the
+  # unprivileged optimizer user cannot write: "Failed to install executable
+  # ... Permission denied". npm/nvm-based harnesses (claude-code, opencode)
+  # are unaffected, so this only bites when the harness changes.
+  UV_TOOL_BIN_DIR: "/home/agent/.local/bin"
 
 wandb:
   project: harness-engineering-bench

harness-engineering-bench/swe-atlas-qna/baseline/build.yaml

@@ -142,4 +142,10 @@ agent_env:
   BASH_DEFAULT_TIMEOUT_MS: "39600000"  # same as max: an un-timed eval must still block
   ENABLE_BACKGROUND_TASKS: "0"
   FORCE_AUTO_BACKGROUND_TASKS: "0"
+  # Harnesses installed with `uv tool install` (mini-swe-agent, swe-agent)
+  # default to symlinking their entry point into /usr/local/bin, which the
+  # unprivileged optimizer user cannot write: "Failed to install executable
+  # ... Permission denied". npm/nvm-based harnesses (claude-code, opencode)
+  # are unaffected, so this only bites when the harness changes.
+  UV_TOOL_BIN_DIR: "/home/agent/.local/bin"
 

harness-engineering-bench/tau3/baseline/build.yaml

@@ -139,6 +139,12 @@ agent_env:
   BASH_DEFAULT_TIMEOUT_MS: "32400000"  # same as max: an un-timed eval must still block
   ENABLE_BACKGROUND_TASKS: "0"
   FORCE_AUTO_BACKGROUND_TASKS: "0"
+  # Harnesses installed with `uv tool install` (mini-swe-agent, swe-agent)
+  # default to symlinking their entry point into /usr/local/bin, which the
+  # unprivileged optimizer user cannot write: "Failed to install executable
+  # ... Permission denied". npm/nvm-based harnesses (claude-code, opencode)
+  # are unaffected, so this only bites when the harness changes.
+  UV_TOOL_BIN_DIR: "/home/agent/.local/bin"
 
 task_environment:
   TAU2_USER_MODEL: openai/gpt-5.4-mini-2026-03-17