Inline orchestration workflow for dependency audit and updates with Beads integration...
You ARE the orchestrator. Execute this workflow directly without spawning a separate orchestrator agent.
Beads Init → Audit → Create Issues → Update by Priority → Close Issues → Verify → Beads Complete
Max iterations: 3 Priorities: critical → high → medium → low Beads integration: Automatic issue tracking
Setup directories:
mkdir -p .tmp/current/{plans,changes,backups}
Validate environment:
package.json existstype-check and build scripts existCreate Beads wisp:
bd mol wisp exploration --vars "question=Dependency audit and update"
IMPORTANT: Save the wisp ID (e.g., mc2-xxx) for later use.
Initialize TodoWrite:
[
{"content": "Dependency audit", "status": "in_progress", "activeForm": "Auditing dependencies"},
{"content": "Create Beads issues", "status": "pending", "activeForm": "Creating issues"},
{"content": "Fix critical dependency issues", "status": "pending", "activeForm": "Fixing critical deps"},
{"content": "Fix high priority dependency issues", "status": "pending", "activeForm": "Fixing high deps"},
{"content": "Fix medium priority dependency issues", "status": "pending", "activeForm": "Fixing medium deps"},
{"content": "Fix low priority dependency issues", "status": "pending", "activeForm": "Fixing low deps"},
{"content": "Verification audit", "status": "pending", "activeForm": "Verifying updates"},
{"content": "Complete Beads wisp", "status": "pending", "activeForm": "Completing wisp"}
]
Invoke dependency-auditor via Task tool:
subagent_type: "dependency-auditor"
description: "Audit all dependencies"
prompt: |
Audit the entire codebase for dependency issues:
- Security vulnerabilities (npm audit / pnpm audit)
- Outdated packages (major/minor/patch)
- Unused dependencies (via Knip)
- Deprecated packages
- License compliance issues
- Categorize by priority (critical/high/medium/low)
Generate: dependency-scan-report.md
Return summary with issue counts per priority.
After dependency-auditor returns:
dependency-scan-report.mdFor each dependency issue found, create a Beads issue:
# Critical security vulnerabilities (P0)
bd create "DEP-SEC: {package}@{version} - {vulnerability}" -t bug -p 0 -d "{description}" \
--deps discovered-from:{wisp_id}
# High - outdated major versions with breaking changes (P1)
bd create "DEP: {package} major update {old} → {new}" -t chore -p 1 -d "{description}" \
--deps discovered-from:{wisp_id}
# Medium - minor updates, deprecated packages (P2)
bd create "DEP: {package} update {old} → {new}" -t chore -p 2 -d "{description}" \
--deps discovered-from:{wisp_id}
# Low - patch updates, unused deps (P3)
bd create "DEP: {package} - {issue}" -t chore -p 3 -d "{description}" \
--deps discovered-from:{wisp_id}
Track issue IDs in a mapping for later closure.
Update TodoWrite: mark "Create Beads issues" complete.
Run inline validation:
pnpm type-check
pnpm build
For each priority (critical → high → medium → low):
Check if issues exist for this priority
Update TodoWrite: mark current priority in_progress
Claim issues in Beads:
bd update {issue_id} --status in_progress
Invoke dependency-updater via Task tool:
subagent_type: "dependency-updater"
description: "Update {priority} dependencies"
prompt: |
Read dependency-scan-report.md and fix all {priority} priority issues.
For each issue:
1. Backup package.json and lockfile
2. Update ONE dependency at a time
3. Run type-check and build after each update
4. If fails, rollback and skip
5. Log change to .tmp/current/changes/deps-changes.json
Generate/update: dependency-updates-implemented.md
Return: count of updated deps, count of failed updates, list of updated dep IDs.
Quality Gate (inline):
pnpm type-check
pnpm build
Close updated issues in Beads:
bd close {issue_id_1} {issue_id_2} ... --reason "Dependency updated"
Update TodoWrite: mark priority complete
Repeat for next priority
After all priorities updated:
Update TodoWrite: mark verification in_progress
Invoke dependency-auditor (verification mode):
subagent_type: "dependency-auditor"
description: "Verification audit"
prompt: |
Re-audit dependencies after updates.
Compare with previous dependency-scan-report.md.
Report:
- Issues fixed (count)
- Issues remaining (count)
- New issues introduced (count)
Decision:
Complete Beads wisp:
# If all updated
bd mol squash {wisp_id}
# If nothing found
bd mol burn {wisp_id}
Create issues for remaining items (if any):
bd create "DEP REMAINING: {package} - {issue}" -t chore -p {priority} \
-d "Not updated in audit. May require manual intervention. See dependency-scan-report.md"
Generate summary for user:
## Dependency Health Check Complete
**Wisp ID**: {wisp_id}
**Iterations**: {count}/3
**Status**: {SUCCESS/PARTIAL}
### Results
- Found: {total} dependency issues
- Fixed: {fixed} ({percentage}%)
- Remaining: {remaining}
### By Priority
- Critical: {fixed}/{total}
- High: {fixed}/{total}
- Medium: {fixed}/{total}
- Low: {fixed}/{total}
### Beads Issues
- Created: {count}
- Closed: {count}
- Remaining: {count}
### Validation
- Type Check: {status}
- Build: {status}
### Artifacts
- Audit: `dependency-scan-report.md`
- Updates: `dependency-updates-implemented.md`
Update TodoWrite: mark wisp complete
SESSION CLOSE PROTOCOL:
git status
git add .
bd sync
git commit -m "chore(deps): {fixed} dependencies updated ({wisp_id})"
bd sync
git push
If quality gate fails:
Rollback available: .tmp/current/changes/deps-changes.json
To rollback:
1. Read changes log
2. Restore package.json and lockfile from .tmp/current/backups/
3. Run pnpm install
4. Re-run workflow
If worker fails:
If Beads command fails:
| Phase | Beads Action |
|---|---|
| 1. Pre-flight | bd mol wisp exploration |
| 3. After audit | bd create for each issue |
| 5. Before update | bd update --status in_progress |
| 5. After update | bd close --reason "Updated" |
| 7. Complete | bd mol squash/burn |
| 7. Remaining | bd create for failed updates |