multi-agent-patterns Eval
=========================

Status: review
Score: 60/100
Risk: medium
Decision: manual_review
Policy: review
Reason: Test manually in an isolated workspace and compare against safer alternatives.

Install:
npx skills add shipshitdev/skills --skill multi-agent-patterns

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
- WARN Agent safety gate: Sparse or mixed signals. Useful for discovery, but not for autonomous installation.
- WARN License clarity: Unknown
- WARN Permission surface: secrets or environment access, filesystem or document access

Warnings:
- Trust score: Potentially useful, but at least one trust signal needs human inspection.
- Audit score: Needs review
- Agent safety gate: Sparse or mixed signals. Useful for discovery, but not for autonomous installation.
- License clarity: Unknown
- Permission surface: secrets or environment access, filesystem or document access
- High-risk permission hints: Secrets or environment access
- License is unclear
- Permission surface may require sandboxing
- plugin.json description is truncated: "Use when single-agent context limits are exceede" is incomplete and inconsistent with SKILL.md.
- SKILL.md uses internal claim IDs such as claim-multi-agent-token-multiplier and claim-evaluation-browsecomp-variance without linking to verifiable sources, making key empirical claims difficult to validate.
- The repository license is detected as Unknown by GitHub; the MIT license is declared in metadata and README, but no LICENSE file is present in the submitted bundle, creating ambiguity for downstream users.
- Low GitHub adoption signal

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
- Low GitHub adoption signal
- plugin.json description is truncated: "Use when single-agent context limits are exceede" is incomplete and inconsistent with SKILL.md.
- High-risk permission hints: Secrets or environment access
- License is unclear
- Permission surface may require sandboxing
- SKILL.md uses internal claim IDs such as claim-multi-agent-token-multiplier and claim-evaluation-browsecomp-variance without linking to verifiable sources, making key empirical claims difficult to validate.

URLs:
- Skill: https://www.openagentskill.com/skills/shipshitdev-multi-agent-patterns
- Audit: https://www.openagentskill.com/skills/shipshitdev-multi-agent-patterns/audit
- JSON: https://www.openagentskill.com/api/agent/evals?slug=shipshitdev-multi-agent-patterns