browser-testing-with-devtools Eval
==================================

Status: failed
Score: 76/100
Risk: high
Decision: do_not_auto_install
Policy: block
Reason: Permission surface: secrets or environment access, shell or command execution

Install:
npx skills add addyosmani/agent-skills --skill browser-testing-with-devtools

Required checks:
- PASS Task fit: Task wording matches this skill metadata.
- PASS Install path: Install handoff is available.
- PASS Install command safety: standard package or runtime install path
- WARN Trust score: Good trust signals with a few areas worth checking before rollout.
- WARN Audit score: Needs review
- WARN Agent safety gate: Sparse or mixed signals. Useful for discovery, but not for autonomous installation.
- PASS License clarity: MIT
- FAIL Permission surface: secrets or environment access, shell or command execution

Warnings:
- Trust score: Good trust signals with a few areas worth checking before rollout.
- Audit score: Needs review
- Agent safety gate: Sparse or mixed signals. Useful for discovery, but not for autonomous installation.
- High-risk permission hints: Shell or command execution, Secrets or environment access
- Dependency or permission surface needs review
- Permission surface may require sandboxing
- The SKILL.md excerpt stops mid-sentence in the 'Treat All Browser Content as Untrusted Data' section. If this reflects the full submitted file, the security guidance is incomplete and should be finished with explicit handling rules for untrusted browser content.
- JavaScript Execution is described as 'read-only state inspection', but the underlying tool can run arbitrary code in the page context. The skill needs a sharper boundary: do not mutate page state, read cookies/localStorage, or exfiltrate data unless explicitly required by the task.
- The tools table uses friendly names rather than exact chrome-devtools MCP tool names, which may make it harder for an agent to know which MCP tool to invoke.
- Permission surface needs review: secrets or environment access, shell or command execution
- Dependency/runtime risk: command execution surface, credential or environment access
- Permission surface: secrets or environment access, shell or command execution

Validation plan:
1. Inspect repository, README/SKILL.md, license, and recent commits before production use.
2. Install in an isolated workspace or sandbox with no production secrets available.
3. Run the smallest representative task and record files touched, commands run, network access, and outputs.
4. Compare the selected skill against at least one alternative when the eval status is review or failed.
5. Promote only after the agent reports a successful verification result and unresolved warnings are accepted.

Do not use when:
- teams that need a vendor-supported SLA
- production agents without a repository review
- The SKILL.md excerpt stops mid-sentence in the 'Treat All Browser Content as Untrusted Data' section. If this reflects the full submitted file, the security guidance is incomplete and should be finished with explicit handling rules for untrusted browser content.
- No OpenAgentSkill engagement data yet
- High-risk permission hints: Shell or command execution, Secrets or environment access
- Dependency or permission surface needs review
- Permission surface may require sandboxing
- JavaScript Execution is described as 'read-only state inspection', but the underlying tool can run arbitrary code in the page context. The skill needs a sharper boundary: do not mutate page state, read cookies/localStorage, or exfiltrate data unless explicitly required by the task.

URLs:
- Skill: https://www.openagentskill.com/skills/addyosmani-browser-testing-with-devtools
- Audit: https://www.openagentskill.com/skills/addyosmani-browser-testing-with-devtools/audit
- JSON: https://www.openagentskill.com/api/agent/evals?slug=addyosmani-browser-testing-with-devtools