一键导入
verification-protocol
How Verifier writes failure reports, including the mandatory Generalizable Learning section that the runtime extracts to learnings/failures/.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
菜单
How Verifier writes failure reports, including the mandatory Generalizable Learning section that the runtime extracts to learnings/failures/.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Complete reference for the factory CLI — all commands, arguments, options, and usage. Use when an agent needs to invoke factory commands, check correct syntax, or understand available options.
When and how to write a human-action-needed entry to memory/{agent}/needs.md. Agents write these entries when they hit blockers that only a human can resolve.
When you encounter a failure, correction, or new pattern, write a learning and optionally propose a skill.
How the factory workspace is organized and what each directory means.
What to examine during a reflection pass and how to write observations to memory/{agent}/needs.md.
When and how to write scenario holdouts for factory verification.
| name | verification-protocol |
| description | How Verifier writes failure reports, including the mandatory Generalizable Learning section that the runtime extracts to learnings/failures/. |
When evaluating a task from tasks/review/:
specs/archive/{task-name}.mdsatisfaction-scoring skill)When verdict is NOT SATISFIED, write to tasks/failed/{task-name}.md:
# Verification Report: {task-name}
spec: specs/archive/{task-name}.md
reviewed_at: {ISO timestamp}
verdict: NOT SATISFIED ({score}/10)
## Summary
{What was evaluated and the overall finding}
## Artifact-by-Artifact Assessment
{For each deliverable in the spec: what was expected, what was found}
## Satisfaction Score: {N}/10
{Reasoning for the score}
## Path to Resolution
{What Builder or a human would need to do to satisfy the spec}
## Generalizable Learning
{REQUIRED — see requirements below}
This section is mandatory in every failure report. The runtime extracts it
to learnings/failures/ automatically. A missing section causes a runtime warning.
The section must:
parser.error() is needed"A useful generalizable learning reads like advice to a future agent who will encounter the same class of situation, not like a description of this specific task.
When a task that previously failed appears in tasks/review/ again: