parallel-monitor Eval
=====================

Status: failed
Score: 62/100
Risk: high
Decision: do_not_auto_install
Policy: block
Reason: Agent safety gate: This skill should not be selected by an agent without explicit human security review.

Install:
npx skills add parallel-web/parallel-agent-skills --skill parallel-monitor

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: Potentially useful, but at least one trust signal needs human inspection.
- WARN Audit score: Needs review
- FAIL Agent safety gate: This skill should not be selected by an agent without explicit human security review.
- PASS License clarity: MIT
- FAIL Permission surface: secrets or environment access, shell or command execution

Warnings:
- Trust score: Potentially useful, but at least one trust signal needs human inspection.
- Audit score: Needs review
- High-risk permission hints: Shell or command execution, Secrets or environment access
- Dependency or permission surface needs review
- Permission surface may require sandboxing
- Financial research output is not financial advice; require human review before any live investment decision
- SKILL.md does not explicitly instruct the agent to treat web-derived monitor event output as untrusted data, leaving room for prompt injection from monitored page content.
- The skill does not include a permission/authorization note reminding the agent to only create monitors for content the user is authorized to track.
- Command construction examples rely on shell quoting but do not mention avoiding direct interpolation of user-controlled queries into shell commands.
- Financial research output is not financial advice; require human review before any live investment decision.
- Quality score needs review
- Permission surface needs review: 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
- SKILL.md does not explicitly instruct the agent to treat web-derived monitor event output as untrusted data, leaving room for prompt injection from monitored page content.
- High-risk permission hints: Shell or command execution, Secrets or environment access
- Dependency or permission surface needs review
- Permission surface may require sandboxing
- Financial research output is not financial advice; require human review before any live investment decision
- The skill does not include a permission/authorization note reminding the agent to only create monitors for content the user is authorized to track.

URLs:
- Skill: https://www.openagentskill.com/skills/parallel-web-parallel-monitor
- Audit: https://www.openagentskill.com/skills/parallel-web-parallel-monitor/audit
- JSON: https://www.openagentskill.com/api/agent/evals?slug=parallel-web-parallel-monitor