gstack/codex/sections/challenge-mode.md

Step 2B: Challenge (Adversarial) Mode

Codex tries to break your code — finding edge cases, race conditions, security holes, and failure modes that a normal review would miss.

1. Construct the adversarial prompt. Always prepend the filesystem boundary instruction from the skill's Filesystem Boundary section (always-loaded skeleton). If the user provided a focus area (e.g., /codex challenge security), include it after the boundary:

Default prompt (no focus): "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 invoke any installed skill (Codex home skills/, .agents/); answer directly. Do NOT modify agents/openai.yaml. Stay focused on repository code only.

Review the changes on this branch against the base branch. Run git diff origin/<base> to see the diff. Your job is to find ways this code will fail in production. Think like an attacker and a chaos engineer. Find edge cases, race conditions, security holes, resource leaks, failure modes, and silent data corruption paths. Be adversarial. Be thorough. No compliments — just the problems."

With focus (e.g., "security"): "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 invoke any installed skill (Codex home skills/, .agents/); answer directly. Do NOT modify agents/openai.yaml. Stay focused on repository code only.

Review the changes on this branch against the base branch. Run git diff origin/<base> to see the diff. Focus specifically on SECURITY. Your job is to find every way an attacker could exploit this code. Think about injection vectors, auth bypasses, privilege escalation, data exposure, and timing attacks. Be adversarial."

2. Create the prompt file and write the full prompt from step 1 into it. Codex reads it on stdin.

_GT="$(git rev-parse --show-toplevel 2>/dev/null || pwd)/.gstack/tmp"
mkdir -p "$_GT" && chmod 700 "$_GT" || { echo "Not sent: cannot create $_GT for the text file." >&2; exit 1; }
_EX=$(git rev-parse --git-path info/exclude 2>/dev/null) && mkdir -p "$(dirname "$_EX")" && { grep -qxF '/.gstack/tmp/' "$_EX" 2>/dev/null || echo '/.gstack/tmp/' >> "$_EX"; }
PROMPT_FILE=$(mktemp "${_GT:?}/codex-prompt.XXXXXX") || { echo "Not sent: mktemp failed in $_GT." >&2; exit 1; }; echo "PROMPT_FILE: $PROMPT_FILE (name: ${PROMPT_FILE##*/})"

Write the text into each printed file with your file-write tool (Claude Code's Write tool needs a Read of the empty file first), exactly as it should appear. The text never goes into a shell command, heredoc or quoted argument. If a write fails or is refused, do not send: print the cause, the file path and the command below for sending by hand.

3. Run codex exec with JSONL output to capture reasoning traces and tool calls. Substitute the printed name for <prompt-file-name> (letters, digits, ., _ and - only). Use timeout: 600000 on the Bash call (the tool's maximum) — the gate sits ABOVE the 540s wrapper so the wrapper fires first, ends Codex, and prints its explicit stall message:

_REPO_ROOT=$(git rev-parse --show-toplevel) || { echo "ERROR: not in a git repo" >&2; exit 1; }
PROMPT_FILE="$(git rev-parse --show-toplevel 2>/dev/null || pwd)/.gstack/tmp/<prompt-file-name>"
[ -s "$PROMPT_FILE" ] || { echo "Not run: $PROMPT_FILE is missing or empty, so the prompt was never written. Write it, then run by hand: codex exec - -C $_REPO_ROOT < $PROMPT_FILE" >&2; exit 1; }
PYTHON_CMD=$(command -v python3 2>/dev/null || command -v python 2>/dev/null || true)
if [ -z "$PYTHON_CMD" ]; then
  echo "ERROR: Python 3 is required to parse Codex JSON output. Install python3 or python and retry." >&2
  exit 1
fi
# Fix 1+2: wrap with timeout (gtimeout/timeout fallback chain via probe helper),
# capture stderr to $TMPERR for auth error detection (was: 2>/dev/null).
[ -n "${TMPERR:-}" ] || TMPERR=$(mktemp "$TMP_ROOT/codex-err-XXXXXX") || { echo "ERROR: mktemp failed in TMP_ROOT=$TMP_ROOT; not running codex without its temp file" >&2; exit 1; }
TMPRESP=$(mktemp "$TMP_ROOT/codex-resp-XXXXXX") || { echo "ERROR: mktemp failed in TMP_ROOT=$TMP_ROOT; not running codex without its temp file" >&2; exit 1; }
_CODEX_PROBE=~/.claude/skills/gstack/bin/gstack-codex-probe
_CODEX_ROLE=''
if [ -n "$_CODEX_ROLE" ]; then export _CODEX_DEADLINE=$(($(date +%s)+540)); _CODEX_OUT=$("$_CODEX_PROBE" role-ready exec) || exit $?
else _CODEX_OUT=$("$_CODEX_PROBE" select-model exec) || exit 1; fi
_CODEX_SEL=$(echo "$_CODEX_OUT" | sed -n 's/^CODEX_SEL: //p')
_CODEX_SANDBOX_MODE=$(echo "$_CODEX_OUT" | sed -n 's/^CODEX_SANDBOX: //p')
"$_CODEX_PROBE" run-with-timeout --user-invoked --effort high --request-effort '<effort-level>' 540 codex exec - -C "$_REPO_ROOT" -s "${_CODEX_SANDBOX_MODE:?}" -c "model=\"${_CODEX_SEL:?}\"" -c skills.include_instructions=false -c 'web_search="cached"' --json -o "$TMPRESP" < "$PROMPT_FILE" 2>"$TMPERR" | tee "$TMPRESP.events" | PYTHONUNBUFFERED=1 "$PYTHON_CMD" -u -c "
import sys, json
turn_completed_count = 0
turn_failed = False
for line in sys.stdin:
    line = line.strip()
    if not line: continue
    try:
        obj = json.loads(line)
        t = obj.get('type','')
        if t == 'item.completed' and 'item' in obj:
            item = obj['item']
            itype = item.get('type','')
            text = item.get('text','')
            if itype == 'reasoning' and text:
                print(f'[codex thinking] {text}', flush=True)
                print(flush=True)
            elif itype == 'agent_message' and text:
                print(text, flush=True)
            elif itype == 'command_execution':
                cmd = item.get('command','')
                if cmd: print(f'[codex ran] {cmd}', flush=True)
        elif t == 'turn.completed':
            turn_completed_count += 1
            usage = obj.get('usage',{})
            tokens = usage.get('input_tokens',0) + usage.get('output_tokens',0)
            if tokens: print(f'\ntokens used: {tokens}', flush=True)
        elif t == 'turn.failed':
            turn_failed = True
            err = obj.get('error',{}).get('message','') or 'no error message in event'
            print(f'[codex turn FAILED] {err}', flush=True, file=sys.stderr)
    except: pass
# Fix 2: three-way completeness check — a STATED failure is a failure,
# not a network problem; only silence with no terminal event is a disconnect.
if turn_failed:
    print('[codex] turn.failed received — the turn errored (reason above), not a disconnect.', flush=True, file=sys.stderr)
elif turn_completed_count == 0:
    print('[codex warning] No turn.completed event received — possible mid-stream disconnect.', flush=True, file=sys.stderr)
"
_CODEX_EXIT=${PIPESTATUS[0]:-${pipestatus[1]}}  # bash sets PIPESTATUS; zsh (lowercase, 1-indexed) falls through
# Fix 1: hang detection — log + surface actionable message
if [ "$_CODEX_EXIT" = "124" ]; then
  "$_CODEX_PROBE" log-event codex_timeout "540"
  "$_CODEX_PROBE" log-hang "challenge" "$(wc -c < "$TMPERR" 2>/dev/null || echo 0)"
  echo "Codex stalled past 9 minutes. Common causes: model API stall, long prompt, network issue. Try re-running. If persistent, split the prompt or check ~/.codex/logs/."
elif [ "$_CODEX_EXIT" != "0" ]; then
  # Surface non-zero exits so the calling agent doesn't read "no output" as
  # a silent model/API stall.
  echo "[codex exit $_CODEX_EXIT] $(head -1 "$TMPERR" 2>/dev/null || echo "no stderr captured")"
  head -20 "$TMPERR" 2>/dev/null | sed 's/^/  /' || true
  "$_CODEX_PROBE" log-event codex_nonzero_exit "challenge:$_CODEX_EXIT"
fi
# Fix 2: surface auth errors from captured stderr instead of dropping them
if grep -qiE "auth|login|unauthorized" "$TMPERR" 2>/dev/null; then
  echo "[codex auth error] $(head -1 "$TMPERR")"
  "$_CODEX_PROBE" log-event codex_auth_failed
fi
bun --no-env-file --config=/dev/null ~/.claude/skills/gstack/lib/outside-review-result.ts --label 'Codex challenge' --exit "$_CODEX_EXIT" --stderr "$TMPERR" --events "$TMPRESP.events" execution "$TMPRESP"
rm -f "$TMPRESP" "$TMPRESP.events" "$PROMPT_FILE"

VERDICT: unavailable means the challenge did not run (its line names why, such as a sandbox that could not start): say so and present no findings from it. Otherwise present the output below.

This parses codex's JSONL events to extract reasoning traces, tool calls, and the final response. The [codex thinking] lines show what codex reasoned through before its answer.

4. Present the full streamed output:

CODEX SAYS (adversarial challenge):
════════════════════════════════════════════════════════════
<full output from above, verbatim>
════════════════════════════════════════════════════════════
Tokens: N | Est. cost: ~$X.XX

4a. Synthesis recommendation (REQUIRED). After presenting the full adversarial output, emit ONE recommendation line summarizing what the user should do, in this format:

Recommendation: <action> because <one-line reason that names the most exploitable finding>

Examples (the strongest reasons compare blast radius across findings or fix-vs-ship):

• Recommendation: Fix the unbounded retry loop Codex flagged at queue.ts:78 because it DoSes the worker pool under sustained 429s, which is higher-blast-radius than the timing leak Codex also flagged that only touches a debug endpoint.
• Recommendation: Ship as-is because Codex's strongest finding is a theoretical race in cleanup that requires conditions we can't trigger in production, weaker than the runtime regressions a fix-now would risk.

The reason must point to a specific finding and compare against alternatives (other findings, fix-vs-ship). Generic reasons like "because it's safer" fail the format. Never silently skip the line.

---