alterlab-boltz Eval
===================

Status: review
Score: 66/100
Risk: medium
Decision: manual_review
Policy: review
Reason: Test manually in an isolated workspace and compare against safer alternatives.

Install:
npx skills add AlterLab-IEU/AlterLab-Academic-Skills --skill alterlab-boltz

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.
- PASS License clarity: MIT
- WARN Permission surface: shell or command execution, database 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.
- Permission surface: shell or command execution, database access
- High-risk permission hints: Shell or command execution
- Financial research output is not financial advice; require human review before any live investment decision
- SKILL.md and references contain unresolved TODO(verify) placeholders for the exact Boltz package pin, YAML/FASTA input schema, and binding-affinity output flag, which are reproducibility blockers.
- evals.json is incomplete: the third eval has an empty expected_output and lacks full assertions.
- Installation and GPU setup instructions are underspecified; no concrete CUDA/torch version or verification steps are provided.
- No example of expected output file names or validation steps is included, making it harder for an agent to confirm success after a run.
- Financial research output is not financial advice; require human review before any live investment decision.
- Quality score needs review

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 and references contain unresolved TODO(verify) placeholders for the exact Boltz package pin, YAML/FASTA input schema, and binding-affinity output flag, which are reproducibility blockers.
- High-risk permission hints: Shell or command execution
- Financial research output is not financial advice; require human review before any live investment decision
- evals.json is incomplete: the third eval has an empty expected_output and lacks full assertions.
- Installation and GPU setup instructions are underspecified; no concrete CUDA/torch version or verification steps are provided.
- No example of expected output file names or validation steps is included, making it harder for an agent to confirm success after a run.

URLs:
- Skill: https://www.openagentskill.com/skills/alterlab-ieu-alterlab-boltz
- Audit: https://www.openagentskill.com/skills/alterlab-ieu-alterlab-boltz/audit
- JSON: https://www.openagentskill.com/api/agent/evals?slug=alterlab-ieu-alterlab-boltz