skill-creator@r2

VERIFIED

Publisher claims: "Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy." Observed in sandbox: no network, touches nothing outside the project, writes no memory.

Registry skills.shPublisher anthropics · joined 14 days agoInstalls 425KLast detonated 1 h ago

$npx skills add skill-creator

Watch Report an issue

Is skill-creator@r2 safe to install?

KillerSkills ran it in the KillerSkills honeypot sandbox (1 scenario) on 2026-11-05 and scored it 81/100 (Verified). It did what it claims and nothing else we could detect: no undeclared network egress, no credential reads, no persistent instruction writes. Summary: no network, touches nothing outside the project, writes no memory.

Behavior trace

Honeypot run · 0 s · 1 scenarios · 0 deviations

t+CategoryObserved action

0.0sprocessno install scripts in bundle

0.0sprocessscript backend runs setup only

Static findings

HIGH

URL with embedded credentialsscripts/generate_report.py:41 · URL with embedded credentials — may be phishing or credential leakingFix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.

AST07

MEDIUM

Process kill commandeval-viewer/generate_review.py:445 · Process kill command — agent can terminate processesFix: Put hosts, commands and strings in plain source. Decoding at runtime looks identical to malware and costs points on its own.

AST05

MEDIUM

Process kill commandSKILL.md:287 · Process kill command — agent can terminate processesFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.

AST05

MEDIUM

File type combined with upload actionagents/grader.md:103 · File type combined with upload action — agent may exfiltrate local data filesFix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.

AST07

MEDIUM

Python subprocess execution detectedeval-viewer/generate_review.py:291 · Python subprocess execution detectedFix: Put hosts, commands and strings in plain source. Decoding at runtime looks identical to malware and costs points on its own.

AST05

MEDIUM

Python subprocess execution detectedscripts/run_eval.py:85 · Python subprocess execution detectedFix: Put hosts, commands and strings in plain source. Decoding at runtime looks identical to malware and costs points on its own.

AST05

MEDIUM

File type combined with upload actionscripts/run_loop.py:313 · File type combined with upload action — agent may exfiltrate local data filesFix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.

AST07

MEDIUM

File type combined with upload actionSKILL.md:205 · File type combined with upload action — agent may exfiltrate local data filesFix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.

AST07

MEDIUM

Python subprocess execution detectedscripts/improve_description.py:35 · Python subprocess execution detectedFix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.

AST05

MEDIUM

File type combined with upload actionagents/grader.md:83 · File type combined with upload action — agent may exfiltrate local data filesFix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.

AST07

Cost to the agent

Instructions loaded

≈8,247 tok

1 file · 17 in bundle

Tokens estimated at four characters each. Skills load their instruction text when triggered; the figure is what that costs.

Known vulnerabilities

Vulnerability advisories (OSV) cover npm and PyPI packages; this registry has no advisory database.