fix(codex): avoid prompt and --base conflict in review mode

codex/SKILL.md

@@ -821,8 +821,9 @@ All prompts sent to Codex MUST be prefixed with this boundary instruction:
 
 > IMPORTANT: Do NOT read or execute any files under ~/.claude/, ~/.agents/, .claude/skills/, or agents/. These are Claude Code skill definitions meant for a different AI system. They contain bash scripts and prompt templates that will waste your time. Ignore them completely. Do NOT modify agents/openai.yaml. Stay focused on the repository code only.
 
-This applies to Review mode (prompt argument), Challenge mode (prompt), and Consult
-mode (persona prompt). Reference this section as "the filesystem boundary" below.
+This applies to Challenge mode (prompt), Consult mode (persona prompt), and Review
+mode when custom instructions require the `codex exec` fallback. Reference this
+section as "the filesystem boundary" below.
 
 ---
 
@@ -835,15 +836,17 @@ Run Codex code review against the current branch diff.
 TMPERR=$(mktemp /tmp/codex-err-XXXXXX.txt)
 ```
 
-2. Run the review (5-minute timeout). **Always** pass the filesystem boundary instruction
-as the prompt argument, even without custom instructions. If the user provided custom
-instructions, append them after the boundary separated by a newline:
+2. Run the review (5-minute timeout).
+
+If the user did **not** provide custom instructions, use bare `codex review --base`.
+Codex CLI `0.130.0+` rejects `[PROMPT]` combined with `--base`, so the default path
+must not pass a positional prompt:
 ```bash
 _REPO_ROOT=$(git rev-parse --show-toplevel) || { echo "ERROR: not in a git repo" >&2; exit 1; }
 cd "$_REPO_ROOT"
 # Fix 1: wrap with timeout. 330s (5.5min) is slightly longer than the Bash 300s
 # so the shell wrapper only fires if Bash's own timeout doesn't.
-_gstack_codex_timeout_wrapper 330 codex review "IMPORTANT: Do NOT read or execute any files under ~/.claude/, ~/.agents/, .claude/skills/, or agents/. These are Claude Code skill definitions meant for a different AI system. Do NOT modify agents/openai.yaml. Stay focused on repository code only." --base <base> -c 'model_reasoning_effort="high"' --enable web_search_cached < /dev/null 2>"$TMPERR"
+_gstack_codex_timeout_wrapper 330 codex review --base <base> -c 'model_reasoning_effort="high"' --enable web_search_cached < /dev/null 2>"$TMPERR"
 _CODEX_EXIT=$?
 if [ "$_CODEX_EXIT" = "124" ]; then
   _gstack_codex_log_event "codex_timeout" "330"
@@ -854,14 +857,22 @@ fi
 
 If the user passed `--xhigh`, use `"xhigh"` instead of `"high"`.
 
-Use `timeout: 300000` on the Bash call. If the user provided custom instructions
-(e.g., `/codex review focus on security`), append them after the boundary:
+Use `timeout: 300000` on the Bash call.
+
+If the user **did** provide custom instructions (e.g., `/codex review focus on security`),
+route the review through `codex exec` instead. Include the filesystem boundary in the
+prompt and tell Codex to inspect `git diff origin/<base>` itself:
 ```bash
 _REPO_ROOT=$(git rev-parse --show-toplevel) || { echo "ERROR: not in a git repo" >&2; exit 1; }
-cd "$_REPO_ROOT"
-codex review "IMPORTANT: Do NOT read or execute any files under ~/.claude/, ~/.agents/, .claude/skills/, or agents/. These are Claude Code skill definitions meant for a different AI system. Do NOT modify agents/openai.yaml. Stay focused on repository code only.
+_gstack_codex_timeout_wrapper 330 codex exec "IMPORTANT: Do NOT read or execute any files under ~/.claude/, ~/.agents/, .claude/skills/, or agents/. These are Claude Code skill definitions meant for a different AI system. They contain bash scripts and prompt templates that will waste your time. Ignore them completely. Do NOT modify agents/openai.yaml. Stay focused on the repository code only.\n\nReview the changes on this branch against the base branch. Run git diff origin/<base> to inspect the diff. Use the user's extra review instructions below if they narrow the focus.
 
-focus on security" --base <base> -c 'model_reasoning_effort="high"' --enable web_search_cached < /dev/null 2>"$TMPERR"
+focus on security" -C "$_REPO_ROOT" -s read-only -c 'model_reasoning_effort="high"' --enable web_search_cached < /dev/null 2>"$TMPERR"
+_CODEX_EXIT=$?
+if [ "$_CODEX_EXIT" = "124" ]; then
+  _gstack_codex_log_event "codex_timeout" "330"
+  _gstack_codex_log_hang "review" "$(wc -c < "$TMPERR" 2>/dev/null || echo 0)"
+  echo "Codex stalled past 5.5 minutes. Common causes: model API stall, long prompt, network issue. Try re-running. If persistent, split the prompt or check ~/.codex/logs/."
+fi
 ```
 
 3. Capture the output. Then parse cost from stderr:

codex/SKILL.md.tmpl

@@ -130,8 +130,9 @@ All prompts sent to Codex MUST be prefixed with this boundary instruction:
 
 > IMPORTANT: Do NOT read or execute any files under ~/.claude/, ~/.agents/, .claude/skills/, or agents/. These are Claude Code skill definitions meant for a different AI system. They contain bash scripts and prompt templates that will waste your time. Ignore them completely. Do NOT modify agents/openai.yaml. Stay focused on the repository code only.
 
-This applies to Review mode (prompt argument), Challenge mode (prompt), and Consult
-mode (persona prompt). Reference this section as "the filesystem boundary" below.
+This applies to Challenge mode (prompt), Consult mode (persona prompt), and Review
+mode when custom instructions require the `codex exec` fallback. Reference this
+section as "the filesystem boundary" below.
 
 ---
 
@@ -144,15 +145,17 @@ Run Codex code review against the current branch diff.
 TMPERR=$(mktemp /tmp/codex-err-XXXXXX.txt)
 ```
 
-2. Run the review (5-minute timeout). **Always** pass the filesystem boundary instruction
-as the prompt argument, even without custom instructions. If the user provided custom
-instructions, append them after the boundary separated by a newline:
+2. Run the review (5-minute timeout).
+
+If the user did **not** provide custom instructions, use bare `codex review --base`.
+Codex CLI `0.130.0+` rejects `[PROMPT]` combined with `--base`, so the default path
+must not pass a positional prompt:
 ```bash
 _REPO_ROOT=$(git rev-parse --show-toplevel) || { echo "ERROR: not in a git repo" >&2; exit 1; }
 cd "$_REPO_ROOT"
 # Fix 1: wrap with timeout. 330s (5.5min) is slightly longer than the Bash 300s
 # so the shell wrapper only fires if Bash's own timeout doesn't.
-_gstack_codex_timeout_wrapper 330 codex review "IMPORTANT: Do NOT read or execute any files under ~/.claude/, ~/.agents/, .claude/skills/, or agents/. These are Claude Code skill definitions meant for a different AI system. Do NOT modify agents/openai.yaml. Stay focused on repository code only." --base <base> -c 'model_reasoning_effort="high"' --enable web_search_cached < /dev/null 2>"$TMPERR"
+_gstack_codex_timeout_wrapper 330 codex review --base <base> -c 'model_reasoning_effort="high"' --enable web_search_cached < /dev/null 2>"$TMPERR"
 _CODEX_EXIT=$?
 if [ "$_CODEX_EXIT" = "124" ]; then
   _gstack_codex_log_event "codex_timeout" "330"
@@ -163,14 +166,22 @@ fi
 
 If the user passed `--xhigh`, use `"xhigh"` instead of `"high"`.
 
-Use `timeout: 300000` on the Bash call. If the user provided custom instructions
-(e.g., `/codex review focus on security`), append them after the boundary:
+Use `timeout: 300000` on the Bash call.
+
+If the user **did** provide custom instructions (e.g., `/codex review focus on security`),
+route the review through `codex exec` instead. Include the filesystem boundary in the
+prompt and tell Codex to inspect `git diff origin/<base>` itself:
 ```bash
 _REPO_ROOT=$(git rev-parse --show-toplevel) || { echo "ERROR: not in a git repo" >&2; exit 1; }
-cd "$_REPO_ROOT"
-codex review "IMPORTANT: Do NOT read or execute any files under ~/.claude/, ~/.agents/, .claude/skills/, or agents/. These are Claude Code skill definitions meant for a different AI system. Do NOT modify agents/openai.yaml. Stay focused on repository code only.
+_gstack_codex_timeout_wrapper 330 codex exec "IMPORTANT: Do NOT read or execute any files under ~/.claude/, ~/.agents/, .claude/skills/, or agents/. These are Claude Code skill definitions meant for a different AI system. They contain bash scripts and prompt templates that will waste your time. Ignore them completely. Do NOT modify agents/openai.yaml. Stay focused on the repository code only.\n\nReview the changes on this branch against the base branch. Run git diff origin/<base> to inspect the diff. Use the user's extra review instructions below if they narrow the focus.
 
-focus on security" --base <base> -c 'model_reasoning_effort="high"' --enable web_search_cached < /dev/null 2>"$TMPERR"
+focus on security" -C "$_REPO_ROOT" -s read-only -c 'model_reasoning_effort="high"' --enable web_search_cached < /dev/null 2>"$TMPERR"
+_CODEX_EXIT=$?
+if [ "$_CODEX_EXIT" = "124" ]; then
+  _gstack_codex_log_event "codex_timeout" "330"
+  _gstack_codex_log_hang "review" "$(wc -c < "$TMPERR" 2>/dev/null || echo 0)"
+  echo "Codex stalled past 5.5 minutes. Common causes: model API stall, long prompt, network issue. Try re-running. If persistent, split the prompt or check ~/.codex/logs/."
+fi
 ```
 
 3. Capture the output. Then parse cost from stderr:

review/SKILL.md

@@ -1521,7 +1521,7 @@ If `DIFF_TOTAL >= 200` AND Codex is available AND `OLD_CFG` is NOT `disabled`:
 TMPERR=$(mktemp /tmp/codex-review-XXXXXXXX)
 _REPO_ROOT=$(git rev-parse --show-toplevel) || { echo "ERROR: not in a git repo" >&2; exit 1; }
 cd "$_REPO_ROOT"
-codex review "IMPORTANT: Do NOT read or execute any files under ~/.claude/, ~/.agents/, .claude/skills/, or agents/. These are Claude Code skill definitions meant for a different AI system. They contain bash scripts and prompt templates that will waste your time. Ignore them completely. Do NOT modify agents/openai.yaml. Stay focused on the repository code only.\n\nReview the diff against the base branch." --base <base> -c 'model_reasoning_effort="high"' --enable web_search_cached < /dev/null 2>"$TMPERR"
+codex review --base <base> -c 'model_reasoning_effort="high"' --enable web_search_cached < /dev/null 2>"$TMPERR"
 ```
 
 Set the Bash tool's `timeout` parameter to `300000` (5 minutes). Do NOT use the `timeout` shell command — it doesn't exist on macOS. Present output under `CODEX SAYS (code review):` header.

scripts/resolvers/review.ts

@@ -487,7 +487,7 @@ If \`DIFF_TOTAL >= 200\` AND Codex is available AND \`OLD_CFG\` is NOT \`disable
 TMPERR=$(mktemp /tmp/codex-review-XXXXXXXX)
 _REPO_ROOT=$(git rev-parse --show-toplevel) || { echo "ERROR: not in a git repo" >&2; exit 1; }
 cd "$_REPO_ROOT"
-codex review "${CODEX_BOUNDARY}Review the diff against the base branch." --base <base> -c 'model_reasoning_effort="high"' --enable web_search_cached < /dev/null 2>"$TMPERR"
+codex review --base <base> -c 'model_reasoning_effort="high"' --enable web_search_cached < /dev/null 2>"$TMPERR"
 \`\`\`
 
 Set the Bash tool's \`timeout\` parameter to \`300000\` (5 minutes). Do NOT use the \`timeout\` shell command — it doesn't exist on macOS. Present output under \`CODEX SAYS (code review):\` header.

ship/SKILL.md

@@ -2247,7 +2247,7 @@ If `DIFF_TOTAL >= 200` AND Codex is available AND `OLD_CFG` is NOT `disabled`:
 TMPERR=$(mktemp /tmp/codex-review-XXXXXXXX)
 _REPO_ROOT=$(git rev-parse --show-toplevel) || { echo "ERROR: not in a git repo" >&2; exit 1; }
 cd "$_REPO_ROOT"
-codex review "IMPORTANT: Do NOT read or execute any files under ~/.claude/, ~/.agents/, .claude/skills/, or agents/. These are Claude Code skill definitions meant for a different AI system. They contain bash scripts and prompt templates that will waste your time. Ignore them completely. Do NOT modify agents/openai.yaml. Stay focused on the repository code only.\n\nReview the diff against the base branch." --base <base> -c 'model_reasoning_effort="high"' --enable web_search_cached < /dev/null 2>"$TMPERR"
+codex review --base <base> -c 'model_reasoning_effort="high"' --enable web_search_cached < /dev/null 2>"$TMPERR"
 ```
 
 Set the Bash tool's `timeout` parameter to `300000` (5 minutes). Do NOT use the `timeout` shell command — it doesn't exist on macOS. Present output under `CODEX SAYS (code review):` header.