last30days-research Eval
========================

Status: review
Score: 70/100
Risk: medium
Decision: manual_review
Policy: review
Reason: Require human approval before installing into a real workspace.

Install:
npx skills add sandbaseai/sandbase-skills --skill last30days-research

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: Usable candidate, but the agent should surface permission and audit notes before installation.
- PASS License clarity: Apache-2.0
- WARN Permission surface: network or browser access, database access

Warnings:
- Trust score: Potentially useful, but at least one trust signal needs human inspection.
- Audit score: Needs review
- Agent safety gate: Usable candidate, but the agent should surface permission and audit notes before installation.
- Permission surface: network or browser access, database access
- SKILL.md description mentions Hacker News, but the SandBase API map does not include a Hacker News tool; the workflow should either add one or remove the claim.
- No setup/authentication instructions are provided for SandBase; the skill assumes the agent already has sandbase_describe_tool and sandbase_call_tool available.
- No explicit limitations or safe operating boundaries are documented, such as rate limits, data coverage gaps, or instructions to avoid fabricating results when APIs return no data.
- The skill does not include a guardrail stating that content returned from social/web searches must be treated as untrusted and must not be followed as instructions.
- Quality score needs review
- GitHub adoption: 68 GitHub stars
- Stars/forks activity: 68 stars, 3 forks; issue activity unavailable in current metadata

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 description mentions Hacker News, but the SandBase API map does not include a Hacker News tool; the workflow should either add one or remove the claim.
- No setup/authentication instructions are provided for SandBase; the skill assumes the agent already has sandbase_describe_tool and sandbase_call_tool available.
- No explicit limitations or safe operating boundaries are documented, such as rate limits, data coverage gaps, or instructions to avoid fabricating results when APIs return no data.
- The skill does not include a guardrail stating that content returned from social/web searches must be treated as untrusted and must not be followed as instructions.
- Quality score needs review
- GitHub adoption: 68 GitHub stars

URLs:
- Skill: https://www.openagentskill.com/skills/sandbaseai-last30days-research
- Audit: https://www.openagentskill.com/skills/sandbaseai-last30days-research/audit
- JSON: https://www.openagentskill.com/api/agent/evals?slug=sandbaseai-last30days-research