agent-runtime-framework/harnesses/contexts/agent-repo/v1/finalize.sh

#!/bin/bash
# Agent repo finalize script — auto-commit and push changes
# AR-19, AR-20, AR-8, AR-32, F66
set -euo pipefail

# ---------------------------------------------------------------------------
# F66: Content-based secret scanning
# Scans staged diff for credential patterns before committing.
# Returns 0 if clean, 1 if secrets detected.
# ---------------------------------------------------------------------------
scan_for_secrets() {
    local diff found=0
    diff=$(cat)

    # Extract only added lines (skip diff headers +++)
    local added
    added=$(printf '%s\n' "$diff" | grep '^+' | grep -v '^+++' || true)

    if [[ -z "$added" ]]; then
        return 0
    fi

    # Pattern 1: AWS access key — AKIA followed by exactly 16 uppercase letters/digits
    if printf '%s\n' "$added" | grep -qE 'AKIA[0-9A-Z]{16}'; then
        echo "AWS access key pattern detected (AKIA...)"
        found=1
    fi

    # Pattern 2: PEM private key header (RSA, EC, OPENSSH, etc.)
    if printf '%s\n' "$added" | grep -qE -- '-----BEGIN .* PRIVATE KEY-----'; then
        echo "PEM private key header detected"
        found=1
    fi

    # Pattern 3: High-entropy credential assignment (40+ char value near key/token/secret keywords)
    if printf '%s\n' "$added" | grep -qiE \
        '(secret[._]?key|api[._]?key|access[._]?key|auth[._]?token|\bsecret\b|\btoken\b|\bpassword\b|\bpasswd\b)[[:space:]]*[=:][[:space:]]*['"'"'""]?[A-Za-z0-9+/=_-]{40,}'; then
        echo "High-entropy credential assignment detected"
        found=1
    fi

    return $found
}

# Allow sourcing for unit tests without running main
if [[ "${BASH_SOURCE[0]}" != "${0}" ]]; then
    return 0 2>/dev/null || true
fi

echo "=== agent-repo/v1 finalize.sh ==="

WORKING_DIR="${AGENT_WORKING_DIR:-/workspace/project}"
AGENT_OUTPUT_DIR="${AGENT_OUTPUT_DIR:-/workspace/.agent-output}"
METADATA_FILE="$AGENT_OUTPUT_DIR/ci_metadata.json"
AGENT_EXIT="${AGENT_EXIT_CODE:-0}"
TASK_ID="${AGENT_TASK_ID:-unknown}"
BRANCH="${AGENT_BRANCH:-}"
REPO_URL="${AGENT_REPO_URL:-}"

mkdir -p "$AGENT_OUTPUT_DIR"

# AR-19: Mark working dir as safe (finalize may run as different user than agent)
git config --global --add safe.directory "$WORKING_DIR"

# Change to working directory
cd "$WORKING_DIR" || {
    echo "ERROR: Failed to change to working directory: $WORKING_DIR" >&2
    exit 1
}

# Verify this is a git repository (defensive: catches misconfigured payloads
# where agent_repo field is missing but agent-repo harness is in the composite)
if [ ! -d ".git" ]; then
    echo "ERROR: $WORKING_DIR is not a git repository." >&2
    echo "Hint: Ensure the task payload includes 'agent_repo: {repo_url: ..., branch: ...}'" >&2
    echo "      so init.sh clones the repo before the agent runs." >&2
    exit 1
fi

# AR-32: Restore .gitignore from harness template to prevent removing entries
HARNESS_GITIGNORE="/opt/harness/contexts/agent-repo/v1/.gitignore_template"
if [ -f "$HARNESS_GITIGNORE" ]; then
    cat "$HARNESS_GITIGNORE" >> .gitignore
fi

# AR-8: Check if there are any changes to commit
git add -A

# AR-37/F73-10: Exclude protected test folders from commit.
# PROTECTED_TEST_FOLDERS is set by entrypoint.py when test_pass_required is active.
# This is defense-in-depth: prevents test file changes reaching the agent branch even
# if the entrypoint revert had a window between the last test run and finalize.
if [ -n "${PROTECTED_TEST_FOLDERS:-}" ]; then
    IFS=',' read -ra _PROTECTED_FOLDERS <<< "$PROTECTED_TEST_FOLDERS"
    for _folder in "${_PROTECTED_FOLDERS[@]}"; do
        _folder=$(echo "$_folder" | xargs)  # trim whitespace
        if [ -n "$_folder" ]; then
            echo "F73: Excluding test folder from commit: $_folder"
            git reset HEAD -- "$_folder" 2>/dev/null || true
            git checkout HEAD -- "$_folder" 2>/dev/null || true
            git clean -fd -- "$_folder" 2>/dev/null || true
        fi
    done
    # Also exclude test infrastructure files
    for _infra in conftest.py pyproject.toml pytest.ini setup.cfg tox.ini; do
        git reset HEAD -- "$_infra" 2>/dev/null || true
        git checkout HEAD -- "$_infra" 2>/dev/null || true
        git clean -fd -- "$_infra" 2>/dev/null || true
    done
fi

if [ -z "$(git status --porcelain)" ]; then
    echo "No changes to commit — skipping push (AR-8)"
    # Write metadata with agent_branch_pushed=false
    python3 -c "
import json, os
out = '$METADATA_FILE'
meta = {}
if os.path.isfile(out):
    try:
        with open(out) as f:
            meta = json.load(f)
    except Exception:
        meta = {}
meta['agent_branch'] = '$BRANCH'
meta['agent_repo_url'] = '$REPO_URL'
meta['agent_branch_pushed'] = False
with open(out, 'w') as f:
    json.dump(meta, f)
print('Wrote ci_metadata.json (no-op: no changes)')
"
    exit 0
fi

# F66: Scan staged content for secrets before committing
SCAN_FINDINGS=""
SCAN_EXIT=0
set +e
SCAN_FINDINGS=$(git diff --cached | scan_for_secrets)
SCAN_EXIT=$?
set -e

if [ "$SCAN_EXIT" -ne 0 ]; then
    echo "SECRET SCAN BLOCKED COMMIT — credentials detected in staged content"
    echo "$SCAN_FINDINGS"
    # Write metadata indicating scan blocked the commit
    python3 -c "
import json, os
out = '$METADATA_FILE'
meta = {}
if os.path.isfile(out):
    try:
        with open(out) as f:
            meta = json.load(f)
    except Exception:
        meta = {}
meta['agent_branch'] = '$BRANCH'
meta['agent_repo_url'] = '$REPO_URL'
meta['agent_branch_pushed'] = False
meta['secret_scan_blocked'] = True
meta['scan_findings'] = '''$SCAN_FINDINGS'''
with open(out, 'w') as f:
    json.dump(meta, f)
print('Wrote ci_metadata.json (blocked by secret scan)')
"
    exit 0
fi

# V3: Diff size check (warning only, does not block push)
DIFF_LINES=$(git diff --cached --stat | tail -1 | grep -oP '\d+(?= insertion)' || echo "0")
DIFF_KB=$(echo "$DIFF_LINES" | awk '{printf "%.1f", $1/1000}')
MAX_DIFF_KB="${AGENT_MAX_DIFF_KB:-500}"
if awk "BEGIN {exit !($DIFF_KB > $MAX_DIFF_KB)}"; then
    echo "WARNING: Large diff detected: ${DIFF_KB}KB (threshold: ${MAX_DIFF_KB}KB)"
fi

# AR-6, AR-18: Build commit message based on agent outcome
if [ "$AGENT_EXIT" = "0" ]; then
    OUTCOME="succeeded"
else
    OUTCOME="failed (exit $AGENT_EXIT)"
fi

# Get first line of prompt (safe, via tempfile)
PROMPT_FIRST_LINE="${AGENT_PROMPT:-}"
if [ -n "$PROMPT_FIRST_LINE" ]; then
    PROMPT_FIRST_LINE=$(echo "$PROMPT_FIRST_LINE" | head -1 | cut -c1-100)
fi

# Write commit message to tempfile (AR-19: safe handling)
COMMIT_MSG_FILE=$(mktemp)
trap "rm -f $COMMIT_MSG_FILE" EXIT

if [ -n "$PROMPT_FIRST_LINE" ]; then
    echo "Agent task $TASK_ID ($OUTCOME): $PROMPT_FIRST_LINE" > "$COMMIT_MSG_FILE"
else
    echo "Agent task $TASK_ID ($OUTCOME)" > "$COMMIT_MSG_FILE"
fi

# Commit
git commit -F "$COMMIT_MSG_FILE"

# Get commit SHA
COMMIT_SHA=$(git rev-parse HEAD)

# AR-19: Always push — even if tests failed, partial work is better than lost work.
# The task exit code and ci_metadata.json track whether tests passed.
echo "Pushing branch $BRANCH to $REPO_URL..."
if timeout 120 git push "$REPO_URL" "HEAD:refs/heads/$BRANCH" --force; then
    echo "Push succeeded"
    PUSHED=true
else
    echo "ERROR: Push failed" >&2
    PUSHED=false
fi

# Write ci_metadata.json
python3 -c "
import json, os
out = '$METADATA_FILE'
meta = {}
if os.path.isfile(out):
    try:
        with open(out) as f:
            meta = json.load(f)
    except Exception:
        meta = {}
meta['agent_branch'] = '$BRANCH'
meta['agent_sha'] = '$COMMIT_SHA'
meta['agent_repo_url'] = '$REPO_URL'
meta['agent_branch_pushed'] = $( [ '$PUSHED' = 'true' ] && echo 'True' || echo 'False' )
meta['diff_kb'] = float('$DIFF_KB') if '$DIFF_KB' else 0.0
with open(out, 'w') as f:
    json.dump(meta, f)
print('Wrote ci_metadata.json')
"

if [ "$PUSHED" = "false" ]; then
    echo "ERROR: Push failed — branch $BRANCH was committed but not pushed" >&2
    exit 1
fi

echo "=== agent-repo/v1 finalize.sh complete ==="