- Home
- Skills
- Agents & Orchestration
- design rereview
Works with the AI Agents you already use
design rereview
Objectively verify architectural fixes and measure project health improvement between design reviews.
$15
design rereview
Example session with this skill installed
Run a re-review for the backend and security lenses against the baseline in docs/remediation/design-review.md. Our current branch is main.
- Read your context and instructions
- Compiled the design rereview
- Generated the document
I have completed the re-review for the
backend and
security lenses. I identified the baseline commit as a1b2c3d and compared it against the current main branch at f4g5h6i.
Design Re-review Report (2023-11-24)
| Lens | Baseline Grade | Current Grade | Change | Findings Fixed | New/Regressed |
|---|---|---|---|---|---|
| Backend | 64 (D) | 82 (B) | +18 | 5 / 7 | 0 |
| Security | 45 (F) | 78 (C) | +33 | 4 / 5 | 1 |
| Overall | 54.5 (F) | 80.0 (B) | +25.5 | 9 / 12 | 1 |
Summary of Changes
Top Improvement: The SQL injection vulnerability (SEC-01) was resolved by migrating to parameterized queries in src/db/gateway.py.
*
Top Improvement: Circular dependencies in the service layer (BACK-03) were eliminated via the new provider pattern.
*
Regression: A new Security finding (SEC-06) was identified in src/auth/oauth.py: hardcoded client secret found in the recent OAuth2 implementation.
*
Score Impediment: The score is currently capped at 'B' because the Database Connection Pool remains unoptimized (BACK-07).
Verified Fixes
BACK-01: Fixed. Logic moved toBaseServiceas seen insrc/services/base.py:L45.SEC-01: Fixed. All raw string interpolations removed fromsrc/db/.SEC-04: Accepted. Decision #12 (Risk accepted for internal dev-only endpoint).
Next steps
- Review the new regression in
src/auth/oauth.pyimmediately. - Run
action-planto address the remaining 3 open findings. - View the full HTML report at
pages/rereview.html.
rereview.html
PDF · document
Example file from a real run - the skill writes it into your workspace.
Connects securely to your tools. The creator never sees your data.
About this skill
Verification of technical debt remediation is often a "he-said, she-said" between developers and auditors. This skill provides an objective, code-backed re-review process to prove that architectural fixes actually landed. It compares current code against a baseline design review to measure score improvements, track regressions, and verify specific remediations.
What it does
- Status tracking verifies every baseline finding as Fixed, Partly Fixed, Still Open, or Regressed.
- Delta analysis scans new and changed code since the baseline for introduced defects.
- Score comparison recomputes grades for both versions to show objective project health movement.
- Decision integration syncs with repo decision records to mark findings as Accepted based on recorded architectural choices.
- Evidence collection mandates specific code citations for every finding marked as Fixed.
How it works
- Identify Baseline by locating the previous
design-review.mdand identifying the specific commit and lenses used. - Scope Changes using
git diffandgit logto map exactly what changed between the baseline and the current branch. - Execute Lenses by running specialized reviewer roles (Architecture, Security, UX, etc.) in parallel to verify findings in their specific domains.
- Compute Comparison using
compare.pyto generate the final scorecard and delta report.
Frameworks & tools
Requires Python 3 for the bundled analysis and comparison scripts. Integrates with standard Git workflows and expects a structured remediation directory (docs/remediation/).
Why this beats prompting it yourself
Generic prompts often suffer from "fix bias" where they assume a ticket closure means a bug is gone. This skill enforces a strict verification protocol that requires proof in the current code and re-runs full grading rubrics to ensure the architectural "score" actually moved.
Use cases
- Post-remediation audit to prove to stakeholders that a security or technical debt sprint achieved its goals.
- Regression checking to ensure that rapid fixes didn't introduce new architectural violations.
- Scorecard reporting to provide quantifiable project health metrics before and after major releases.
Known limitations
Requires a valid baseline design-review.md to function. Statuses for runtime or production-only issues are marked as Not Verifiable if the proof is not present in the source code.
Requires an existing design-review.md baseline. Verification is limited to source code; runtime or production-only behaviors are marked as Not Verifiable.
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