- Home
- Skills
- Agents & Orchestration
- action plan
Works with the AI Agents you already use
action plan
Convert design reviews and audits into sequenced remediation roadmaps with risk scores and rollback plans.
$12
action plan
Example session with this skill installed
Read the design-review.md in docs/remediation and create an action plan. We have a release freeze in two weeks and low capacity for the backend team.
- Read your context and instructions
- Compiled the action plan
I have processed the findings from docs/remediation/design-review.md. I applied your constraints regarding the two-week release freeze and limited backend capacity. High-priority containment for the security findings is scheduled for this week, while backend structural changes are deferred to the hardening phase.
Remediation Action Plan Summary
| ID | Finding | Phase | Priority | Effort | Owner |
|---|---|---|---|---|---|
| A-001 | F-SEC-001: SQL Injection risk in auth | Containment | Critical | Low | Security |
| A-002 | F-BE-004: N+1 Query in dashboard | Near-term | High | Med | Backend |
| A-003 | F-UX-002: Missing ARIA labels | Hardening | Med | Low | Frontend |
| A-004 | F-ARC-001: Monolith service split | Long-term | Low | High | Arch |
Containment Plan (Week 1)
- A-001 (Critical): Implement input validation middleware on the auth endpoint.
- Rollback: Revert to commit
HEAD^1; existing DB state is unaffected. - Observability: Monitor 403 Forbidden spikes in CloudWatch.
- Acceptance: Snyk scan passes on
/api/auth.
- Rollback: Revert to commit
Decision Log (Blockers)
- D-001: Stakeholder approval needed for A-004 service split due to infrastructure costs.
- D-002: Confirm if PII data encryption (A-005) requires a maintenance window for backfilling.
Next steps
- Review the full plan at
docs/remediation/action-plan.md. - Run
plan-to-issuesto sync these tasks to GitHub. - Approve the drafted ADRs in
docs/remediation/decision-drafts/.
Connects securely to your tools. The creator never sees your data.
About this skill
When a design review or security audit lands on your desk, the gap between "knowing what is wrong" and "knowing what to do on Monday" is a major source of friction. This skill bridge that gap by converting raw findings into a sequenced remediation plan. It moves from immediate containment to long-term hardening, ensuring every critical vulnerability or architectural flaw is mapped to a specific, actionable item.
What it does
- Remediation sequencing organizes fixes into containment, near-term, hardening, and long-term phases.
- Traceability mapping links every finding ID (F-SEC-001) to specific action IDs (A-001) for audit compliance.
- Risk scoring calculates effort, priority, and impact for every work item.
- Rollback planning enforces recovery and validation steps for every data migration or system change.
- Decision logging identifies blockers that require stakeholder sign-off before implementation can start.
- ADR drafting generates Architectural Decision Records for proposed changes in your repo's specific format.
How it works
- Input ingestion reads your
design-review.mdand any conflict reports from previous audit stages. - Constraint collection prompts for your specific SLOs, deadlines, capacity, and risk tolerance.
- Plan generation invokes the planner to validate findings against code and write the
action-plan.mdartifact. - Quality gating verifies that every high-severity finding is addressed and every action includes observability and rollback steps.
- Issue preparation summarizes the first-week actions and offers to export items to GitHub.
Frameworks & tools
Built for environments using GitHub for issue tracking. The automation scripts require Python 3. It integrates with optional companion tools for code mapping and design memory.
Why this beats prompting it yourself
Generic prompts often overlook the "how" of a fix, such as rollback procedures or data validation. This skill enforces a strict quality gate that prevents "hand-wavy" remediation, ensuring every task is ready for an engineer to pick up.
Use cases
- Post-audit remediation after a security or architectural review.
- Technical debt reduction by turning a backlog of findings into a structured roadmap.
- Incident response planning where containment must happen before long-term fixes.
Known limitations
Requires Python 3 for bundled scripts. Does not automatically create GitHub issues without the companion plan-to-issues skill. Planning quality depends on the detail in the input review report.
Requires Python 3 for scripts. Does not auto-create GitHub issues without plan-to-issues. Depends on structured findings input.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 12 days ago
Requires