agent-browser
73
92%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
View guide
Critical
Do not install without reviewing
Security
2 findings: 1 critical severity, 1 high severity. Installing this skill is not recommended: please review these findings carefully if you do intend to do so.
E005: Suspicious download URL detected in skill instructions.
What this means
Detected a suspicious URL in the skill instructions that could lead the agent to download and execute malicious scripts or binaries. This includes links to executables from untrusted sources, typosquatting of official packages, URL shorteners that obscure the destination, and personal file hosting services.
Why it was flagged
While many entries are benign placeholders or official pages (example.com, lightpanda.io docs, github.com/login), the list also contains an explicit malicious domain (malicious.com) and several ambiguous/unvetted hosts (site-a.com, site-b.com, prod/staging/app subdomains) and includes localhost/GitHub login entries — because the skill instructs downloading/executing from these URLs the presence of an explicitly malicious domain and untrusted hosts makes the set a suspicious potential malware-distribution vector.
W007: Insecure credential handling detected in skill instructions.
What this means
The skill handles credentials insecurely by requiring the agent to include secret values verbatim in its generated output. This exposes credentials in the agent’s context and conversation history, creating a risk of data exfiltration.
Why it was flagged
The prompt includes examples that embed plaintext passwords and session tokens directly in CLI commands and state files (e.g., agent-browser fill "password123", state files with session tokens), which would require an LLM to include secrets verbatim in generated commands or code.
Low
Low-risk findings.
1 low severity finding. Worth noting, but not necessarily harmful.
W011: Third-party content exposure detected (indirect prompt injection risk).
What this means
The skill exposes the agent to untrusted, user-generated content from public third-party sources, creating a risk of indirect prompt injection. This includes browsing arbitrary URLs, reading social media posts or forum comments, and analyzing content from unknown websites.
Why it was flagged
This skill explicitly automates browsing arbitrary URLs and ingesting page content (e.g., SKILL.md core workflow and templates show agent-browser open <url>, agent-browser snapshot -i, and agent-browser get text body) so the agent will read untrusted third-party web content and use it to drive subsequent actions.
- Repository
- tibelf/ai_project_init
- Commit
56f4516
- Audited
- Security analysis
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.