Skip to content

fix: jira-agent posts /auto-cc to wrong PR due to head -1 URL extraction #74612

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
openshift-merge-bot merged 1 commit into from
Feb 10, 2026
Merged

fix: jira-agent posts /auto-cc to wrong PR due to head -1 URL extraction #74612

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
55 changes: 53 additions & 2 deletions ...tor/step-registry/hypershift/jira-agent/process/hypershift-jira-agent-process-commands.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -225,8 +225,59 @@ while IFS= read -r line; do
echo "Claude processing complete. Full output saved to /tmp/claude-${ISSUE_KEY}-output.json"

if [ $EXIT_CODE -eq 0 ]; then
# Parse PR URL from result if available
PR_URL=$(echo "$RESULT" | grep -oP 'https://github.com/openshift/hypershift/pull/[0-9]+' | head -1 || echo "")
# Extract the PR URL created by Claude. Strategy 1 (primary): parse the
# structured JSONL output to find tool_result content for "gh pr create"
# Bash tool calls. This is deterministic and avoids extra GitHub API calls.
# Strategy 2 (fallback): verify candidate URLs via gh pr view (costs N API
# calls but handles schema changes or missing JSONL output).
PR_URL=""
OUTPUT_FILE="/tmp/claude-${ISSUE_KEY}-output.json"

if [ -f "$OUTPUT_FILE" ]; then
# Extract tool_use IDs for "gh pr create" Bash commands.
# grep '^{' skips non-JSON stderr lines (from 2>&1).
IDS=$(grep '^{' "$OUTPUT_FILE" | jq -rs '
[.[] |
select(.type == "assistant") |
.message.content[]? |
select(.type == "tool_use" and .name == "Bash" and
(.input.command | tostring | contains("gh pr create"))) |
.id
]' 2>/dev/null || echo '[]')

# If we found gh pr create calls, get the PR URL from matching results
if [ "$IDS" != "[]" ] && [ -n "$IDS" ]; then
PR_URL=$(grep '^{' "$OUTPUT_FILE" | jq -rs --argjson ids "$IDS" '
[.[] |
select(.type == "user") |
.message.content[]? |
select(.type == "tool_result" and (.tool_use_id | IN($ids[]))) |
.content |
if type == "array" then .[].text // empty
elif type == "string" then .
else empty end
] |
map(capture("(?<url>https://github\\.com/[^/]+/[^/]+/pull/[0-9]+)").url) |
last // empty
' 2>/dev/null || echo "")
fi
fi

# Fallback: verify candidate URLs via gh pr view if JSONL parsing found nothing
if [ -z "$PR_URL" ]; then
CANDIDATE_URLS=$(echo "$RESULT" | grep -oP 'https://github.com/openshift/hypershift/pull/[0-9]+' | sort -u || true)
if [ -n "$CANDIDATE_URLS" ]; then
while IFS= read -r url; do
PR_JSON=$(gh pr view "$url" --json title,author 2>/dev/null || true)
PR_TITLE=$(echo "$PR_JSON" | jq -r '.title // empty' 2>/dev/null || true)
PR_AUTHOR=$(echo "$PR_JSON" | jq -r '.author.login // empty' 2>/dev/null || true)
if [[ "$PR_TITLE" == "${ISSUE_KEY}"* ]] && [[ "$PR_AUTHOR" == *"hypershift-jira-solve-ci"* ]]; then
PR_URL="$url"
break
fi
done <<< "$CANDIDATE_URLS"
fi
fi

@celebdor celebdor Feb 9, 2026

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Wouldn't it be better to look only at the gh pr create entries?:

      # Pass 1: Get tool_use_ids for "gh pr create" commands
      IDS=$(jq -rs '
        [.[] |
         select(.type == "assistant" and .message.content) |
         .message.content[] |
         select(.type == "tool_use" and .name == "Bash" and (.input.command | contains("gh pr create"))) |
         .id
        ] | @json
      ' /tmp/stream-test.jsonl)

      # Pass 2: Get PR URLs and store in bash array
      mapfile -t PR_URLS < <(jq -rs --argjson ids "$IDS" '
        [.[] |
         select(.type == "user" and .message.content) |
         .message.content[] |
         select(.type == "tool_result" and (.tool_use_id | IN($ids[]))) |
         .content
        ] |
        map(capture("(?<url>https://github.com/[^/]+/[^/]+/pull/[0-9]+)").url) |
        .[]
      ' /tmp/stream-test.jsonl)

      # Demonstrate the array
      echo "Number of PRs created: ${#PR_URLS[@]}"
      echo "PR URLs array contents:"
      for i in "${!PR_URLS[@]}"; do
        echo "  [$i] ${PR_URLS[$i]}"
      done)


echo "✅ Successfully processed $ISSUE_KEY"
if [ -n "$PR_URL" ]; then
Expand Down