Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

snyk-scan · szewowsky | DecimalAI

securityby szewowsky

snyk-scan

szewowsky/snyk-scan

Source szewowsky/warsztat-skille-cc ↗License MIT

Skanuje skille, paczki MCP i agentów pod kątem zagrożeń (prompt injection, kradzież kluczy, złośliwy kod) przez Snyk agent-scan. Używaj gdy: „przeskanuj skilla", „sprawdź czy skill jest bezpieczny", „zeskanuj paczkę" — dla lokalnego folderu lub linku do repo ze skillem.

SkillScore86Maintained

Automatic

Security & AppSec pack

Last verified 8/3/2026

What it changes

Passed 19 of 22 test cases with the skill, and 12 of 22 without it.

Without skill55%

With skill86%

Checks it passed that the model alone missed

Classifies financial access under the Medium severity level.
Classifies system modifications under the Medium severity level.
Specifies `snyk-agent-scan@latest` as the package invoked by uvx.

and 4 more in the full report

7 fixed12 passed either way3 still failing0 broke

Pass-rate lift: +32 pp
Tokens: -10%
Agent turns: 0%
Response time: -44%

Model gemini-3.6-flashMeasured 8/3/2026v119 of 22 cases passed with skill

Tokens, turns and time describe resource use, not a dollar price.

Use this skill

Save it into your project's skills folder. Claude Code picks it up in its next session.

curl -fsSL --create-dirs -o .claude/skills/szewowsky-snyk-scan/SKILL.md https://app.decimal.ai/s/szewowsky-snyk-scan/SKILL.md

Want it pinned and updatable? pip install "decimalai>=0.15", then decimalai skills add szewowsky-snyk-scan, and start Claude Code with claude --plugin-dir "$(git rev-parse --show-toplevel)/.agents"

1 · One-time — install the CLI (docs ↗)

pip install "decimalai>=0.15"

2 · Get the skill — writes .agents/skills/szewowsky-snyk-scan/ and pins it in .decimal/skills.lock

decimalai skills add szewowsky-snyk-scan

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/szewowsky-snyk-scan/SKILL.md https://app.decimal.ai/s/szewowsky-snyk-scan/SKILL.md

Raw file: /s/szewowsky-snyk-scan/SKILL.md ↗ · machine-readable summary at /s/szewowsky-snyk-scan.json ↗

Safety checks: 3 of 3 passed

Code scan CleanIntent PassedContent Clean

See it in action

One of 7 test cases the skill fixed. Same model, same prompt, with and without the skill.

1 of 7

Prompt

A skill requests permission to read crypto wallet keys and payment configurations. Which severity level categorizes financial access risks?

What was checked, without and with the skill
Checked	Without skill	With skill
Classifies financial access under the Medium severity level.	Missed	Passed

case-14 · gemini-3.6-flash · 8/3/2026 · answers are capped and redacted

What's inside

The instructions your model reads when the skill loads.

SKILL.md37 lines2 filesincludes scripts

snyk-scan

Sprawdza czy skill z internetu nie próbuje Cię okraść, zanim go zainstalujesz. Silnik: uvx snyk-agent-scan@latest (Snyk). To bezpieczeństwo, nie higiena (od higieny opisu jest osobny audyt skilli).

Wymagania (jednorazowo)

uv zainstalowany (brew install uv).
SNYK_TOKEN ustawiony w środowisku — z https://app.snyk.io/account (API Token → KEY). Bez tokena skan nie ruszy.

Jak uruchomić

Deterministyczna logika (rozpoznanie URL vs ścieżka, klonowanie, skan, sprzątanie) siedzi w skrypcie — NIE zgaduj, odpal go:

bash ~/.claude/skills/snyk-scan/scripts/scan.sh "<CEL>"

gdzie <CEL> to:

lokalny folder ze skillem, np. ~/Downloads/pdf-optimizer — skanuje w miejscu,
link do repo, np. https://github.com/mattpocock/skills — skrypt sam klonuje do katalogu tymczasowego, skanuje i po skanie go usuwa.

Po skanie streść użytkownikowi wynik: liczba i kategorie znalezisk (Critical / High / Medium) + werdykt „bezpieczny / uważaj / nie instaluj".

Zasady

Zawsze podawaj jawny CEL. NIGDY nie odpalaj uvx snyk-agent-scan@latest bez ścieżki — bez celu narzędzie skanuje wszystkie lokalne skille użytkownika (wolno, szum).
Wymaga internetu (pobranie narzędzia + serwer weryfikacji).
Nie instaluj skanowanego skilla dopóki nie jest czysty.

Kategorie wykrywane przez Snyk

Critical: prompt injection, malicious code
High: credential handling, secret detection
Medium: third-party content, unverifiable dependencies, financial access, system modifications

Track record

Benchmark

19 / 22

test cases passed, vs 12 / 22 without

Adoption

Users: 0
Times used: 0

Safety and source details

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/1/2026

View 15 scanner checks

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

szewowsky / warsztat-skille-cc

snyk-scan/SKILL.md

View on GitHub ↗

License MITImported 8/1/2026Version v1Bundled 1 file · scripts

More skills from the same source and nearby in the registry, with their headline benchmark lift.

Ready to supercharge your AI agents?

See it in action

What's inside

snyk-scan

Wymagania (jednorazowo)

Jak uruchomić

Zasady

Kategorie wykrywane przez Snyk

Track record

Benchmark

Adoption

SkillSafety

Source

Related skills