Solo Create Issue

Solo Create Issue

Parse the user's request and create a GitHub issue in hiero-ledger/solo with full project metadata using one Bash tool call that runs all GraphQL mutations sequentially in a single script. All IDs are pre-resolved — no lookup calls needed.

Step 1 — Parse the user's request

Extract the following from the user's natural-language description:

Body Format

## Overview
<2-4 sentence summary of the goal and motivation>

## Sub-Tasks
- [ ] Sub-task one
- [ ] Sub-task two

## Prerequisites
- [ ] Any blockers or dependencies (omit section if none)

### Dependencies
- [ ] requires #N    ← issues this one depends on
- [ ] required by #N ← issues that depend on this one

## References
- [Link text](url) — one line per reference (omit section if none)

Step 2 — Resolve IDs from the tables below

Repository (fixed)

Issue Types

Solo X Team Project

Status Options

Priority Options

Labels

Priority → Label mapping (always include the priority label)

Step 3 — Execute all API calls in a single Bash tool call

Combine all calls into one script and run as a single Bash tool invocation — one approval prompt instead of many.

The calls are sequential and chain outputs:

1. Create issue → capture issue.id and issue.url (type + labels set in this same mutation)
2. Add to project using issue.id → capture item.id
3. Set Status + Priority using item.id (both fields in one aliased mutation)
4. (Optional) Link as sub-issue if the user specified a parent epic/initiative

set -euo pipefail

TITLE="<synthesized title>"
BODY=$(cat <<'BODY_EOF'
<formatted body>
BODY_EOF
)

# ── Call 1: Create issue (type + labels included) ─────────────────────────────
RESPONSE1=$(gh api graphql -f query="
mutation {
  createIssue(input: {
    repositoryId: \"R_kgDOLMTWdQ\"
    title: $(echo "$TITLE" | jq -Rs .)
    issueTypeId: \"<ISSUE_TYPE_ID>\"
    labelIds: [\"<LABEL_ID_1>\"]
    body: $(echo "$BODY" | jq -Rs .)
  }) {
    issue { number url id }
  }
}")
ISSUE_ID=$(echo "$RESPONSE1"  | jq -r '.data.createIssue.issue.id')
ISSUE_URL=$(echo "$RESPONSE1" | jq -r '.data.createIssue.issue.url')

# ── Call 2: Add to Solo X Team project ───────────────────────────────────────
RESPONSE2=$(gh api graphql -f query="
mutation {
  addProjectV2ItemById(input: {
    projectId: \"PVT_kwDOCq2Q984A6EW6\"
    contentId: \"$ISSUE_ID\"
  }) {
    item { id }
  }
}")
ITEM_ID=$(echo "$RESPONSE2" | jq -r '.data.addProjectV2ItemById.item.id')

# ── Call 3: Set Status + Priority (aliased — one round trip) ──────────────────
gh api graphql -f query="
mutation {
  setStatus: updateProjectV2ItemFieldValue(input: {
    projectId: \"PVT_kwDOCq2Q984A6EW6\"
    itemId: \"$ITEM_ID\"
    fieldId: \"PVTSSF_lADOCq2Q984A6EW6zguwhjU\"
    value: { singleSelectOptionId: \"<STATUS_OPTION_ID>\" }
  }) { projectV2Item { id } }

  setPriority: updateProjectV2ItemFieldValue(input: {
    projectId: \"PVT_kwDOCq2Q984A6EW6\"
    itemId: \"$ITEM_ID\"
    fieldId: \"PVTSSF_lADOCq2Q984A6EW6zguwhkA\"
    value: { singleSelectOptionId: \"<PRIORITY_OPTION_ID>\" }
  }) { projectV2Item { id } }
}" > /dev/null

# ── Call 4 (optional): Link as sub-issue of parent epic/initiative ────────────
# Uncomment and set PARENT_NODE_ID when user specifies a parent.
# Get it via: gh api graphql -f query='{ repository(owner:"hiero-ledger",name:"solo") { issue(number:N) { id } } }'
#
# PARENT_NODE_ID="<parent issue node ID>"
# gh api graphql -f query="
# mutation {
#   addSubIssue(input: {
#     issueId: \"$PARENT_NODE_ID\"
#     subIssueId: \"$ISSUE_ID\"
#     replaceParent: false
#   }) { issue { number } subIssue { number } }
# }" > /dev/null

echo "$ISSUE_URL"

Step 4 — Report result

Created: <issue url>
Title:    <title>
Type:     <type>  |  Priority: <P0/P1/P2>  |  Status: <status>
Labels:   <label names>
Project:  Solo X Team

Tips from the field

• Batch creation: when creating many issues (e.g. populating an epic), capture all issue IDs first in Pass 1, then do a second pass to patch cross-reference links (- [ ] requires #N / - [ ] required by #N) into each body once all numbers are known.

• Status guidance: use Ready for issues with no blockers; use Backlog when the issue depends on other issues being completed first.

• Parent node IDs for common epics/initiatives (save a lookup):

  Issue	#	Node ID
  2026 Q3 — Developer Experience	5004	I_kwDOLMTWdc8AAAABIo7dFw
  2026 Q3 — User Experience	5002	I_kwDOLMTWdc8AAAABIoWEfQ
  2026 Q3 — Technical Debt	5018	I_kwDOLMTWdc8AAAABIqZ7jQ