一键导入
finding-format
Required fields and severity rubric (CRITICAL/HIGH/MEDIUM/LOW) for review findings, with file:line evidence convention. Owned by the reviewer.
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
Required fields and severity rubric (CRITICAL/HIGH/MEDIUM/LOW) for review findings, with file:line evidence convention. Owned by the reviewer.
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
Keep commits atomic — split a mixed change set into scoped, reviewable commits.
The one-time procedure to take an exported project to live — understand it, confirm the name/goal/stack with the user, run the deterministic `aspis bootstrap`, enrich the judgment files (AGENTS.md, ARCHITECTURE, context), verify, and let the package self-clean. Followed by the bootstrap agent.
Keep per-subsystem architectural intent current through the planning loop — read before designing, record an impact report on change, confirm with the user, apply a dated update, and verify the build against approved intent.
Audit a plan's task dependency graph for structural integrity — circular dependencies, missing prerequisites, orphan tasks, and dependency classification (hard/soft/warning). Produces a pass/warn/fail audit report per dependency so planners catch graph errors before build starts.
Every editing agent should start on a clean working tree so parallel work never collides.
Verify hooks ran, no secrets, protected paths untouched, and commit message valid before committing. Owned by the reviewer.
| name | finding-format |
| description | Required fields and severity rubric (CRITICAL/HIGH/MEDIUM/LOW) for review findings, with file:line evidence convention. Owned by the reviewer. |
Standardize every reviewer's finding so it is machine-parseable, actionable, and carries enough evidence for the builder to fix the issue without re-investigation. Eliminates vague output like "this looks wrong" and grounds every severity call in the rubric defined in the F-016 reviewer reference spec §6 (the canonical verdict system — cited, not restated, in this skill).
Every time the reviewer produces a finding — plan review (plan-critic),
change review (quality-review, security-review), scope-compliance check, or
acceptance verdict (acceptance-decision). If the agent is the reviewer, the
finding runs through this format.
For every issue found, capture the six required fields in this order:
file:line reference (repo-relative path, line
number; use path:start-end for a range).CRITICAL → always REJECTED · HIGH → always CHANGES-REQUIRED ·
MEDIUM → blocking by default, deferrable only with an explicit note ·
LOW → never blocking, always a note.git diff HEAD~1:L138-L150),
gate failure message, or rule ID. "No evidence = no verdict" — if any
field cannot be filled, withhold the verdict and request the missing
evidence from the delegating lead.A finding with all six fields above, ready to fold into the review report and route to the build lead (HIGH / MEDIUM-blocking) or committer (LOW notes).
file:line — the builder cannot locate the issue.