One row per merged PR, appended with ship log <pr> … right after the merge. The retro reads it to decide what to keep: which step found what, what each PR cost, and what escaped to main.
claude-skill
security-control-testing
DRiskyibragimov-oasis/name-navigator
Technical verification procedures for ISO 27002 control assessment.
security-control-testing got grade D on our automated scan on 2026-11-03.
Why this grade
- Has a line over 4000 characters, too long to scan reliably.
- Contains invisible or bidi Unicode (U+061C).
- Tells the agent to skip permissions or safety checks.
- Tells the agent to skip permissions or safety checks.
- Tells the agent to skip permissions or safety checks.
- Tells the agent to skip permissions or safety checks.
- Tells the agent to skip permissions or safety checks.
- Some files were not scanned (too many, too big, or not text), so the grade is capped at C.
Before you install
- Read any scripts this skill ships.
- Note every URL it fetches.
- Check allowed-tools and hooks.
Embed this grade
Add this badge to your README. It shows the current grade from CheckSkills.
Report this skill if the grade looks wrong, or if something slipped through.
Related
"""Generate comprehensive index metadata report"""
规则被组织为一个通用层加上语言特定的目录：
URL: https://www.youtube.com/watch?v=87Ybz1pE2F4 Views: 1663 Date: 20260903 Length: 5:39 ADDRESSES GOAL: Yes, provides real usage data, speed/cost numbers, prompting strategies for long agentic runs, and comparisons with GPT-5.6 Sol. HANDS-ON: Yes (OpenCode terminal CLI harness connected via API).
set -eu