菜单
Evaluate a decision against the user's stated principles + related past thinking. Loads principles from YAML, matches by keyword, surfaces semantically similar prior decisions.
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。
Raw accounting of what's in the brain. Counts messages, embeddings, summaries, domains. Use when user asks "how much is in my brain" or "what's the size of the corpus."
Surface patterns in how the user thinks — cognitive pattern, problem-solving approach, emotional tone, content category. Optional domain scope. Use when user asks "when do I think best" or "what patterns show up in breakthroughs."
Full re-entry brief for a domain — recent summaries + accumulated open questions, decisions, insights, quotes. Longer and deeper than tunnel-state. Use after days/weeks away from a project.
Browse conversations by calendar date. Use when user asks "what was I working on last Thursday" or specifies a date.
Surface abandoned domains with unresolved breakthrough/significant open questions. Ranked by breakthrough count → question count → conversation count. Use when user feels important threads have been dropped.
Fetch the full message thread of a specific conversation by ID. Returns role + content + timestamp, truncated at 20 messages / 1000 chars per message. Use when the user references a specific conversation from prior search results.
基于 SOC 职业分类
| name | alignment-check |
| description | Evaluate a decision against the user's stated principles + related past thinking. Loads principles from YAML, matches by keyword, surfaces semantically similar prior decisions. |
| layer | L3 |
| reads | ["principles.yaml","L1.embeddings","L0.all_conversations"] |
| writes | [] |
| citations | required |
| determinism | temporal |
| allowed-tools | mcp__my-brain__alignment_check |
L3 fusion skill that joins two different substrates: the user's stated principles (YAML file) + their past thinking (L1 vector search). Produces an evaluation surface for a pending decision. The user interprets — this skill surfaces the relevant material.
decision: str # the decision under consideration, natural-language
## 🎯 Alignment Check: "{decision (truncated 80ch)}"
### Relevant principles (top 3)
- **{principle_name}**: {definition truncated 200ch}
*core formula: {formula}*
- ...
### Related past thinking (top 5)
- [{date}] **{conversation_title}** [conv_id]
> {300ch preview}
### Surface, do not decide
The user reconciles principles with past thinking. This skill does not return a verdict.
min_sim=0.35 (stricter than tunnel-state's 0.3 — fewer but higher-quality past-thinking matches).