codex-memory-guard Eval
=======================

Status: failed
Score: 61/100
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 wl1650918245/codex-memory-guard --skill codex-memory-guard

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
- The workflow depends on scripts/initialize-memory-system.ps1, scripts/install.ps1, and scripts/test.ps1, but these scripts are not included in the submitted skill file listing. The skill package should include them or clearly link to a stable repository path so users can run and audit the full workflow.
- SKILL.md instructs reading HANDOFF.md and memory files at task start but does not explicitly state that memory file contents must be treated as data, not as instructions. This is worth hardening against prompt injection from a compromised or malicious memory file.
- Low GitHub adoption signal
- Quality score needs review
- Permission surface needs review: secrets or environment access, shell or command execution
- GitHub adoption: 22 GitHub stars
- Stars/forks activity: 22 stars, 1 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
- Low GitHub adoption signal
- The workflow depends on scripts/initialize-memory-system.ps1, scripts/install.ps1, and scripts/test.ps1, but these scripts are not included in the submitted skill file listing. The skill package should include them or clearly link to a stable repository path so users can run and audit the full workflow.
- High-risk permission hints: Shell or command execution, Secrets or environment access
- Dependency or permission surface needs review
- Permission surface may require sandboxing
- SKILL.md instructs reading HANDOFF.md and memory files at task start but does not explicitly state that memory file contents must be treated as data, not as instructions. This is worth hardening against prompt injection from a compromised or malicious memory file.

URLs:
- Skill: https://www.openagentskill.com/skills/wl1650918245-codex-memory-guard
- Audit: https://www.openagentskill.com/skills/wl1650918245-codex-memory-guard/audit
- JSON: https://www.openagentskill.com/api/agent/evals?slug=wl1650918245-codex-memory-guard