Why task approvals and recycle bin keep governance fast
Without approval, errors ship. With heavy approval, delivery stops. Task approvals plus recycle bin (soft delete) balance control and undo.
WKFGo: TASK_APPROVAL, RecycleBinPage, DeletedTask, restore_task.
Symptoms and cost
| Signal | Cost |
|---|---|
| Approval in email | No audit |
| Delete = fear | Zombie cards |
| Unbounded queue | Everything waiting |
| Reject without feedback | Endless rework |
Common scenarios
Ten-person product team. Needs visibility without daily meetings—live board and decision wiki.
Bilingual agency. Persian stakeholders and English devs—FA/EN UI and Jalali calendar are non-negotiable.
Enterprise-light. Roles, approvals, audit—without a half-time Jira admin.
A practical Task Approvals and Recycle Bin framework
What needs approval?
Legal, budget, release—not every bug.
Name an owner and set a weekly deadline. Output must be visible on the board, wiki, or reports—verbal “done” in standup is not enough for audit. If the step stalls, record the blocker on the task, not only in chat.
Approver roles
ProjectRole + TASK_APPROVAL.
Name an owner and set a weekly deadline. Output must be visible on the board, wiki, or reports—verbal “done” in standup is not enough for audit. If the step stalls, record the blocker on the task, not only in chat.
Queue SLA
flow_aging on approvals.
Name an owner and set a weekly deadline. Output must be visible on the board, wiki, or reports—verbal “done” in standup is not enough for audit. If the step stalls, record the blocker on the task, not only in chat.
Reject requires comment
TaskHistory.
Name an owner and set a weekly deadline. Output must be visible on the board, wiki, or reports—verbal “done” in standup is not enough for audit. If the step stalls, record the blocker on the task, not only in chat.
RecycleBin policy
restore_task within 30 days.
Name an owner and set a weekly deadline. Output must be visible on the board, wiki, or reports—verbal “done” in standup is not enough for audit. If the step stalls, record the blocker on the task, not only in chat.
Weekly queue drain
TaskApprovalsPage.
Name an owner and set a weekly deadline. Output must be visible on the board, wiki, or reports—verbal “done” in standup is not enough for audit. If the step stalls, record the blocker on the task, not only in chat.
Email vs in-app
| Channel | Audit |
|---|---|
| Weak | |
| Task approval | Strong |
Trend metrics (4–6 weeks)
| Signal | Why |
|---|---|
| % tasks updated weekly | Real adoption |
| Status meeting duration | Should shrink |
| Blockers >3 days | Bottleneck |
| Decisions logged in wiki | Organizational memory |
Anti-patterns
- Approve everything
- Approver without backup
- Hard delete default
- Silent reject
How WKFGo helps
TaskApprovalsPage; decide_approval MCP; RecycleBinPage; FeatureAccess TASK_APPROVAL.
Pilot wrap-up
Pick one real project this week—not a sandbox. Name a PM or lead owner. Ninety minutes: diagnose (which table row?), execute two framework steps, log one decision in wiki or decision_inbox. Next week check whether status meetings shortened or blockers surfaced faster.
Fifteen-minute retro after pilot: what worked? which gap remains? Log the decision in wiki so the next lead does not restart from zero.
WKFGo applies JWT and FeatureAccess consistently across UI, MCP, and chatbot—test with a contractor account before org-wide rollout.
Compare monthly reports as trends—not one-day snapshots. If metrics stall, the bottleneck is usually policy or capacity.
FA/EN teams must test UI and Jalali calendar in real demos—Persian stakeholders lose trust with wrong Gregorian dates.
One team behavior should change—PRs with task IDs, wiki after meetings, or weekly approval queue drain.
FAQ — Task Approvals and Recycle Bin
Multiple approvers?
Per-project policy.
MCP?
list_pending_approvals, decide_approval.
Soft delete?
DeletedTask + restore.