Install
$ agentstack add skill-adeerkhan-vitruvius-peer-review ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
Engineering Peer Review
Run an independent adversarial review of an engineering artifact. This skill reuses the engineering-research evidence discipline with a review-specific lens. The reviewer has NOT seen the author's reasoning — independence is the point.
Invocation
/peer-review
Attach or point to: research brief, design document, calculation set, or report.
Methodology
- Identify artifact and claims — what is it, what does it claim, what standard/code governs it? List every critical claim, number, and citation.
- Verify load-bearing items — for each critical claim:
- Trace to source (standard + section + line, or URL)
- Re-derive any calculation from stated formula, inputs, and units
- Check code/standard applicability (right domain? right edition? right jurisdiction?)
- Confirm the source actually supports the claim at the pinned location
- Produce severity-graded findings:
- FATAL — unsafe, non-compliant, or unsupported; must fix before use
- MAJOR — likely wrong or a real gap; fix before finalizing
- MINOR — clarity, consistency, polish
- Each finding cites the specific section/line/claim it targets
- Check what was NOT verified — mark
blockedwhere the source could not be read. Never let a missing check read as a pass.
- Write a revision plan — prioritized, concrete steps per finding.
Output
Inline Summary (chat response)
- Artifact scope and claim count
- FATAL / MAJOR / MINOR counts
- Top 3 issues with evidence trails
- Overall revision risk assessment
Full Review (saved to disk)
Save to outputs/peer-review/.md:
# Peer Review:
## Summary Assessment
- **Artifact type:**
- **Claims reviewed:**
- **FATAL:** | **MAJOR:** | **MINOR:**
- **Revision risk:** low / medium / high
## Strengths
-
## Findings
### FATAL
#### F1:
- **Location:**
- **Issue:**
- **Evidence:**
- **Fix:**
### MAJOR
#### M1:
- **Location:**
- **Issue:**
- **Evidence:**
- **Fix:**
### MINOR
#### m1:
- **Location:**
- **Issue:**
- **Suggestion:**
## Verification Status
- **Verified:** claims traced to source
- **Blocked:** claims (source unreachable)
- **Unverified:** claims (default, not yet checked)
## Revision Plan
1.
2.
3.
4.
## Evidence Trail (Line-Pinned)
| # | Finding | Source | Location (§/line) | Verdict |
|---|---------|--------|-------------------|---------|
| 1 | | | | contradicts / partially supports / missing |
Scope and Boundaries
- This skill identifies issues — it does NOT approve or sign off. Final engineering sign-off requires a licensed engineer.
- Research-only, not for final engineering sign-off.
- Default-FAIL posture: every claim starts unverified, must be traced to source.
- Never soften a FATAL to MAJOR to avoid being harsh. Engineering accountability requires honest verdicts.
- Evidence quality: see
references/evidence-quality-tiers.md.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: adeerkhan
- Source: adeerkhan/vitruvius
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.