selvage · Nerlo

selvage

By: Selvage Team
License: Apache-2.0
Last scan: Scanned Aug 22, 2026, 8:31 AM UTC

Claude Skill

Flagged31composite

LLM 기반 코드 리뷰 도구

Claude Skills are markdown + scripts loaded into Claude Code; scanned with the same pipeline.

Source:: github.com/selvage-lab/selvage
License:: Apache-2.0
Author:: Selvage Team
Provenance:: Found by crawling (provenance, not a security rating)

We found this by crawling public sources and nobody has attested to it. That is how most of the registry gets here, and it is not a mark against the artifact — it only means no authenticated act of publication is on record. An artifact can be the vendor's own and still appear here.

7 critical findings reported by trivy, osv-scanner. The composite score alone does not reflect this — read the per-scanner reports below.

Install

Claude Skills install to a single target, no runtime picker.

nerlo install -- selvage

Manual alternative: copy the skill folder into ~/.claude/skills/ so Claude Code loads it on next launch.

cp -r -- selvage ~/.claude/skills/

Install respects the composite badge: Clean proceeds, Caution prompts for confirmation, and Flagged is refused. Write operations require an API token.

How scores work Full audit report

Per-scanner reports

11 scanners

Your view:

8 of 11 scanners examined this package. The rest do not apply to it — language, ecosystem and packaging decide which analyzers can say anything, and a scanner that cannot examine an artifact reports nothing rather than passing it.

Scanners that examined this package (8)

agentshield
v1.4.0
Clean100
- No findings
0 findings · 1.0s
cisco-skill-scanner
v2.0.11
Unavailable
Incomplete run
0 findings · 21.8s
agent-audit-kit
v0.3.74
Flagged8
- 9High
- 1Low
10 findings · 26.8s

AI review of these findings

Findings in files this artifact installs stand as reported; a model re-read the rest and said which ones it believes are false positives. This is a second opinion published beside the evidence, not a correction to it: the per-scanner reports above are unchanged, every dismissed finding is still listed there at its original severity, and the score and the Flagged badge are computed from those raw severities alone. Nothing below moved them.

Upheld as real: 308
Dismissed as false positives: 4
Not reviewed: 0

Out of 312 findings across every scanner that examined this package. None of the 7 findings rated critical by their scanner were dismissed.

Why those were dismissed

Two keys had to turn. The model had to name a reason from a fixed list of six, and a deterministic check of ours — over which files this artifact actually installs — had to independently agree. Anything it could not corroborate is in “not reviewed” above, at full severity.

2 Test or fixture file — The match is in a test or fixture. Our own check confirmed the file is not part of what gets installed.
Rules: AAK-SECRET-004 · Files: tests/test_usage_response.py, e2e/e2e_realistic_workflows.py · reported by agent-audit-kit
1 Agent instruction file — The match is in a file that instructs an AI coding agent — a CLAUDE.md or similar. It is documentation of intent, and our own check confirmed it is not part of what gets installed.
Rules: AAK-AGENT-003 · Files: CLAUDE.md · reported by agent-audit-kit
1 Documentation or example — The match is in documentation or sample code, confirmed by our own check to be outside the installed artifact.
Rules: opt.nerlo-rules.nerlo-install-pipe-to-shell · Files: /repo/docs/uv_배포_가이드.md · reported by nerlo-install-instruction

In installed code — never dismissed

These findings are in files this artifact installs and are not placeholder values. No false-positive basis can be corroborated for such a finding, so this review never dismisses one; they were not sent to the model and stand at their scanner's severity.

4 AAK-SECRET-004
Files: selvage/src/utils/llm_client_factory.py, selvage/src/llm_gateway/claude_gateway.py, selvage/src/utils/token/token_utils.py · reported by agent-audit-kit
4 opt.nerlo-rules.nerlo-install-pipe-to-shell
Files: /repo/README.md, /repo/README_KR.md · reported by nerlo-install-instruction
2 GHSA-2f96-g7mh-g2hx
Files: /repo/uv.lock, uv.lock · reported by osv-scanner, trivy
2 GHSA-3f7w-8rr8-f37f
Files: uv.lock, /repo/uv.lock · reported by trivy, osv-scanner
2 GHSA-3rp5-jjmw-4wv2
Files: uv.lock, /repo/uv.lock · reported by trivy, osv-scanner
2 GHSA-4gmw-gg2m-w46p
Files: uv.lock, /repo/uv.lock

Upheld — the review declined to dismiss these

The model could have called any of these a false positive and did not.

1 CVE-2026-54282
Files: uv.lock · reported by trivy
1 CVE-2026-69248
Files: uv.lock · reported by trivy

What the review concluded

Artifact contains critical vulnerabilities and exposed secrets.

The artifact includes dependencies with critical vulnerabilities that could lead to remote code execution or denial of service. It also contains hardcoded secrets in source files, which could grant unauthorized access to external services. Additionally, installation instructions suggest piping to shell, which can execute arbitrary commands.

Concern level assessed by the review: Critical. This is the review's scale, not the registry badge.

What to look at

PYSEC-2026-2102 · /repo/uv.lock · critical · reported by osv-scanner
PYSEC-2026-2161 · /repo/uv.lock · critical · reported by osv-scanner
PYSEC-2026-287 · /repo/uv.lock · critical · reported by osv-scanner
PYSEC-2026-338 · /repo/uv.lock · critical · reported by osv-scanner
PYSEC-2026-36 · /repo/uv.lock · critical · reported by osv-scanner
CVE-2026-27962 · uv.lock · critical · reported by trivy
CVE-2026-32871 · uv.lock · critical · reported by trivy
AAK-MARKETPLACE-001 · .claude-plugin/marketplace.json · high · reported by agent-audit-kit
AAK-SECRET-004 · selvage/src/llm_gateway/claude_gateway.py · high · reported by agent-audit-kit

Reviewed by gemini-2.5-flash. The model sees the findings, not your code, and cannot change a score.

Scan history

1 scan on record. Every scan's full results are retained immutably for 24 months.

Completed	Composite	Change	Scanners	Status
Aug 22, 2026	Flagged31	—	agentshield: completecisco-skill-scanner ·!: incompleteagent-audit-kit: completenerlo-behavioral: completenerlo-install-instruction: completecapslock ·!: incompletenerlo-multi-source ·n/a: not applicabletrivy: completeosv-scanner: completetrivy_image ·n/a: not applicablegovulncheck ·n/a: not applicable	completed

Per-scanner history:

AI review of these findings

Upheld as real: 308
Dismissed as false positives: 4
Not reviewed: 0

Out of 312 findings across every scanner that examined this package. None of the 7 findings rated critical by their scanner were dismissed.

Why those were dismissed

2 Test or fixture file — The match is in a test or fixture. Our own check confirmed the file is not part of what gets installed.
Rules: AAK-SECRET-004 · Files: tests/test_usage_response.py, e2e/e2e_realistic_workflows.py · reported by agent-audit-kit
1 Agent instruction file — The match is in a file that instructs an AI coding agent — a CLAUDE.md or similar. It is documentation of intent, and our own check confirmed it is not part of what gets installed.
Rules: AAK-AGENT-003 · Files: CLAUDE.md · reported by agent-audit-kit
1 Documentation or example — The match is in documentation or sample code, confirmed by our own check to be outside the installed artifact.
Rules: opt.nerlo-rules.nerlo-install-pipe-to-shell · Files: /repo/docs/uv_배포_가이드.md · reported by nerlo-install-instruction

In installed code — never dismissed

4 AAK-SECRET-004
Files: selvage/src/utils/llm_client_factory.py, selvage/src/llm_gateway/claude_gateway.py, selvage/src/utils/token/token_utils.py · reported by agent-audit-kit
4 opt.nerlo-rules.nerlo-install-pipe-to-shell
Files: /repo/README.md, /repo/README_KR.md · reported by nerlo-install-instruction
2 GHSA-2f96-g7mh-g2hx
Files: /repo/uv.lock, uv.lock · reported by osv-scanner, trivy
2 GHSA-3f7w-8rr8-f37f
Files: uv.lock, /repo/uv.lock · reported by trivy, osv-scanner
2 GHSA-3rp5-jjmw-4wv2
Files: uv.lock, /repo/uv.lock · reported by trivy, osv-scanner
2 GHSA-4gmw-gg2m-w46p
Files: uv.lock, /repo/uv.lock

Upheld — the review declined to dismiss these

The model could have called any of these a false positive and did not.

1 CVE-2026-54282
Files: uv.lock · reported by trivy
1 CVE-2026-69248
Files: uv.lock · reported by trivy

What the review concluded

Artifact contains critical vulnerabilities and exposed secrets.

Concern level assessed by the review: Critical. This is the review's scale, not the registry badge.

What to look at

PYSEC-2026-2102 · /repo/uv.lock · critical · reported by osv-scanner
PYSEC-2026-2161 · /repo/uv.lock · critical · reported by osv-scanner
PYSEC-2026-287 · /repo/uv.lock · critical · reported by osv-scanner
PYSEC-2026-338 · /repo/uv.lock · critical · reported by osv-scanner
PYSEC-2026-36 · /repo/uv.lock · critical · reported by osv-scanner
CVE-2026-27962 · uv.lock · critical · reported by trivy
CVE-2026-32871 · uv.lock · critical · reported by trivy
AAK-MARKETPLACE-001 · .claude-plugin/marketplace.json · high · reported by agent-audit-kit
AAK-SECRET-004 · selvage/src/llm_gateway/claude_gateway.py · high · reported by agent-audit-kit

Reviewed by gemini-2.5-flash. The model sees the findings, not your code, and cannot change a score.

selvage

Install

Per-scanner reports

Scanners that examined this package (8)

agentshield

cisco-skill-scanner

agent-audit-kit

nerlo-skill-scanner (behavioral)

nerlo-skill-scanner (install-doc)

capslock

trivy

osv-scanner

Scanners that did not run on this package (3)

nerlo-multi-source

trivy (image)

govulncheck

AI review of these findings

Why those were dismissed

In installed code — never dismissed

Upheld — the review declined to dismiss these

What the review concluded

What to look at

Scan history

AI review of these findings

Why those were dismissed

In installed code — never dismissed

Upheld — the review declined to dismiss these

What the review concluded

What to look at

Install

Per-scanner reports

Scanners that examined this package (8)

agentshield

cisco-skill-scanner

agent-audit-kit

nerlo-skill-scanner (behavioral)

nerlo-skill-scanner (install-doc)

capslock

trivy

osv-scanner

Scanners that did not run on this package (3)

nerlo-multi-source

trivy (image)

govulncheck

Composite scoreOne view among many — the per-scanner reports above are primaryFlagged31

AI review of these findings

Why those were dismissed

In installed code — never dismissed

Upheld — the review declined to dismiss these

What the review concluded

What to look at

Scan history

AI review of these findings

Why those were dismissed

In installed code — never dismissed

Upheld — the review declined to dismiss these

What the review concluded

What to look at