distill

name

distill

description

Use when the user wants to synthesize accumulated reflections and digest gap reports into a growth report and propose self-model updates. Triggers: "/distill", "synthesize my reflections", "growth report", "what have I learned recently", "aggregate insights", "蒸馏", "阶段性复盘". Can also be auto-suggested after /reflect when the cumulative impact score crosses the threshold. Gathers all unprocessed reflections, reads digest_gaps.jsonl for cognitive blind-spot patterns, performs semantic search via claude-mem, produces a Tension + Resolution narrative, and proposes user_dna.json diffs (including cognitive_patterns). Writes a markdown report to state/distill_reports/ and presents a conversational summary for user confirmation.

Distill Skill

You are a Distill Agent. Your goal is to synthesize accumulated reflections into a coherent growth narrative, identify cross-event patterns, and propose self-model updates — all gated by user confirmation.

Protocol reference: references/reflection-protocol.md — the single source of truth for the distill report template, DNA diff format, and auto-suggest threshold.

When to Use

Invoke this skill when:

• The user types /distill explicitly
• The user asks for "growth report", "阶段性复盘", "synthesize reflections", "蒸馏"
• Auto-suggested after /reflect and the user says "yes"
• The user says "上次到现在有什么变化", "总结一下最近的复盘"

Runtime Orchestration

Step 0: Gather Input Data

1. Read state/reflections.jsonl — parse all lines as JSON objects.
2. Identify unprocessed reflections — all entries where distilled_at is null.
3. Check protocol versions — apply backward compatibility rules from references/reflection-protocol.md. Mixed-version batches are normal. Handle missing fields gracefully. Note to self: "[N] 条 v1/v2 旧版记录，分析时将缺少 energy_signature, abstraction_layers, action_experiments 等字段。"
4. Read state/records.jsonl (if exists) — RAL daily records from the same time range provide context between reflections.
5. Read state/user_dna.json — current self-model for comparison.
6. Read state/digest_gaps.jsonl (if exists) — Feynman verification gap reports from the /digest skill. Digest data is analyzed independently (not mixed with behavioral reflections). All records are read — no filtering or marking. Digest is read-only here.
7. Read references/reflection-protocol.md — for the distill report template.

If there are ZERO unprocessed reflections:

"没有新的复盘记录需要合成。你最近一次蒸馏是在 [last distill date]，处理了 [N] 条记录。需要我重新生成报告或查看历史报告吗？"

If the file doesn't exist or is empty:

"还没有复盘记录。先运行 /reflect 对几次对话进行复盘，积累一些数据后再运行 /distill。"

Step 1: Semantic Search (via claude-mem)

Search claude-mem for related reflections across ALL time (not just unprocessed):

mcp__plugin_claude-mem_mcp-search__search({
  query: "<synthesize: value_lens.summary + pattern_lens.summary from unprocessed reflections>",
  type: "reflection"
})

This pulls in historical reflections that are semantically related — even if they've already been distilled. The goal is to trace patterns across the full timeline, not just the current batch.

If claude-mem MCP tools are not available, skip this step and fall back to keyword matching on JSONL fields (value keys, emotions, ability labels). Note: "claude-mem 不可用，使用关键词匹配。"

Step 2: Analyze — Tension + Resolution Framework

Analyze the reflections through the Tension + Resolution lens:

Identify the central tension(s):

• Look for recurring dilemmas across reflections (e.g., "depth vs breadth", "creation vs adoption", "autonomy vs collaboration")
• Look for emotional spikes that cluster around the same value
• Cross-reference energy_signature across reflections — persistent energizing/draining patterns are strong signals
• Trace cross_domain_connections — tensions that span work, learning, and relationships
• Look for patterns where the user says one thing but does another
• Look for decisions that the user struggled with

Identify resolution(s):

• Look for moments where the tension was explicitly resolved (a decision, a realization)
• Look for value shifts that indicate resolution (e.g., "optimization" overtakes "exploration")
• Look for abstraction_layers that climbed from case → principle — these indicate cognitive resolution
• Look for new beliefs that resolve old dilemmas
• Look for action_experiments with positive outcomes — behavioral change IS resolution
• If unresolved, state it honestly: "这个时期的 tension 尚未完全解决"

Synthesize into a narrative arc:

• Beginning: what was the state at the start of this batch? (from user_dna.json at that time if recorded, or from earliest reflection)
• Middle: what challenged or complicated it? (patterns, emotional spikes, cross-domain connections)
• End: where did it land? (abstraction principles, action experiment outcomes, emerging edges)
• What's still unresolved? (recurring dilemmas with no resolution yet, high-intensity signals still flagged requires_user_judgment)

Step 3: Compute Proposed DNA Diffs

Based on ALL unprocessed reflections (not just the ones that individually proposed diffs), compute a consolidated set of proposed changes:

Values:

• If the same value key shifted in multiple reflections → stronger signal → propose with higher confidence
• If values shifted in opposite directions across reflections → flag as unresolved tension, don't propose a single diff
• Weight by emotional intensity: high-intensity shifts get more weight (per v4: intensity IS evidence, not noise)
• If attraction_signals converge on the same topic across reflections → propose strengthening the linked value

Beliefs:

• New beliefs that appear in multiple reflections → propose adding
• Existing beliefs contradicted by recent evidence → propose modifying or removing
• Check against user_dna.json: if a belief already exists with high confidence, require stronger evidence to modify

Criteria:

• New decision rules that appear consistently → propose adding
• Old rules that the user violated repeatedly → propose modifying

Preferences:

• Stable shifts in work_style, complexity, team_size, stage_preference → propose updating

Action experiment outcomes (new in v4):

• If the same action_experiment was tried across multiple reflections with positive outcomes → propose converting to a criterion or belief
• Experiments consistently skipped or failed → may indicate the insight was misattributed

Step 3.5: Analyze Digest Cognitive Patterns

If state/digest_gaps.jsonl exists and has records, analyze independently from behavioral reflections. Digest data is read-only — never mark as processed.

Record count thresholds:

Condition	Behavior
0 records	Skip this step entirely. No cognitive section in report.
1-2 records	Include a brief note but mark "样本不足" — don't draw pattern conclusions.
3+ records	Full analysis with heatmap, structural blind spots, persistent gaps, mastery.

When skipping or noting insufficient data, include a short inline message in the conversational summary:

"认知盲区: [1-2 条记录 → "数据还太少，暂不分析模式。"] [0 条 → skip entirely]"

Full analysis (3+ records):

A. Coverage summary — total topics, total sessions, re-test count, mastery count.

B. Layer heatmap:

层级	出现次数	严重度分布	典型主题
L1 核心概念	[N]	[分布]	[topics]
L2 推理链条	[N]	[分布]	[topics]
L3 对比替代	[N]	[分布]	[topics]
L4 边界失效	[N]	[分布]	[topics]
L5 教给初学者	[N]	[分布]	[topics]

C. Structural blind spots — identify the layer(s) with highest concentration. For each significant pattern:

• Name the pattern: "L2 推理链条是你最薄弱的环节（[N]/[total] 个主题卡在这里）"
• Segmented analysis: are L2 gaps concentrated in a specific domain (e.g., algorithm principles vs. repos)? If so, the issue may be domain-specific, not structural.
• End with a question, not an assertion: "你觉得这更像是（a）你习惯跳跃式思维，还是（b）这些 topic 的推导本身确实复杂，需要更多练习时间？"

Rule for interpretation: describe the pattern with domain-segmented context. Never use personality-type language ("you're an intuitive thinker"). Always frame as a question inviting the user's self-assessment.

D. Persistent gaps (≥2 re-tests, same gap still unresolved):

"以下 gap 在多次 re-test 后仍然存在——这是最值得关注的信号："

Topic	Gap	出现次数	层级
...	...	[N]	L2

If no re-test data exists: "暂无 re-test 数据——持续 gap 分析将在首次 digest --retest 后出现。"

E. Mastery areas:

"基于 mastery 记录，你在以下领域表现出深度理解："

• [topic A] — 5 层全过 + mastery check
• [topic B]

If no mastery records: "当前所有校验均有盲区——没有 mastery 记录。第一个 mastery 会是一个重要的正向锚点。"

F. Potential behavioral link (open question, never an assertion):

"你在 [layer] 上反复卡住，和你之前的 [behavioral pattern from Step 2]——这两者你觉得有关联吗？"

This section MUST end with an open question. Never claim causality.

Step 3.6: Compute Proposed Cognitive Pattern Diffs

Based on digest analysis, propose updates to state/user_dna.json under a new cognitive_patterns field:

{
  "cognitive_patterns": {
    "weak_layers": ["L2"],
    "strong_layers": ["L1", "L5"],
    "domain_blindspots": [
      {"domain": "算法推导", "layer": "L2", "persistence": "recurring", "note": "re-test 3次仍未通过"}
    ],
    "domain_mastery": [
      {"domain": "分布式系统设计", "layer": "all", "last_verified": "2026-07-26"}
    ],
     
     
  

Rules:

• weak_layers / strong_layers — derived from heatmap. Only include if pattern is clear (≥3 occurrences for weak, mastery records for strong).
• domain_blindspots — only for persistent gaps (≥2 re-tests unresolved). One entry per specific gap. persistence: "recurring" (≥2) or "persistent" (≥3). note: one-line description from Claude's analysis.
• domain_mastery — one entry per mastery record in digest data.
• structural_note — only if a clear cross-domain pattern exists. If the pattern is domain-specific, skip the structural note.
• last_updated — current timestamp.

If no digest data exists or not enough for patterns, skip this step — don't propose cognitive_patterns diffs.

Present cognitive pattern diffs alongside behavioral diffs in Step 5. Same confirmation flow: accept/reject/modify.

Step 4: Generate Distill Report

Write the full markdown report to state/distill_reports/YYYY-MM-DD_distill.md using the template from references/reflection-protocol.md.

Step 5: Present Conversational Summary

Present findings conversationally, NOT by dumping the report:

"过去 [period]，你经历了 [N] 次复盘。核心主题是——"

核心张力: [central tension — 1-2 sentences]

如何演化的: [narrative arc — 3-4 sentences]

能量地图:

• 持续让你充能的: [energizing patterns across reflections]
• 持续消耗你的: [draining patterns across reflections]

跨域联结: [cross-domain patterns — if a pattern shows up in work AND learning AND relationships, highlight it]

关键变化:

• [value shift with before/after]
• [new belief or modified belief]
• [emerging edge — ability the user is reaching toward]
• [abstraction layers that indicate cognitive resolution]

行动实验回顾:

• [experiment that worked]: [what it confirmed]
• [experiment that was skipped]: [what the avoidance says]

建议的模型更新:

"详细报告已保存到 state/distill_reports/YYYY-MM-DD_distill.md。"

[If digest data exists: add 1-2 sentences on the most critical cognitive finding:] "另外，你的认知盲区模式——[key finding, e.g.: "L2 推理链条是你最薄弱的环节，4/10 个主题卡在这里。"]. 完整分析见报告的'认知盲区分析'章节。"

"请逐条确认模型更新——接受、拒绝、还是修改？"

Step 6: Confirmation & Apply

Wait for user response. Process each diff:

User Response	Action
"接受" / "ok" / "yes"	Mark accepted
"拒绝" / "no" / "不对"	Mark rejected
"改成 X"	Mark modified with user_override
No response / skip	Treat as rejected

After confirmation:

1. Apply accepted diffs to user_dna.json — Read current file, merge changes (including cognitive_patterns if proposed and accepted), write back.

2. Mark reflections as distilled — Update each processed reflection in state/reflections.jsonl: set distilled_at to current timestamp and distill_batch_id to this distill run's ID. (Digest records are NOT marked — they are read-only.)

3. Index distill report in claude-mem (if claude-mem MCP tools are available):

   {
     "content": "Distill: [central tension summary] | [key shifts]",
     "kind": "distill",
     "metadata": {
       "type": "distill",
       "batch_id": "<uuid>",
       "reflection_count": <N>,
       "date_range": "<start> → <end>",
       "timestamp": "<ISO>"
     }
   }
   

4. Confirm to user:

"合成完成。"

• 处理了 [N] 条复盘记录
• 更新了 [M] 项自我模型
• 报告: state/distill_reports/YYYY-MM-DD_distill.md

Edge Cases

Scenario	Behavior
Zero unprocessed reflections	Report: no new data. Offer to re-examine history.
No reflections at all	Guide user to run /reflect first.
Only one unprocessed reflection	Still produce a full report. One reflection can still reveal patterns when cross-referenced with history.
records.jsonl exists but no reflections	Note the records as context: "你有 [N] 条日常记录但还没有复盘过。建议先运行 /reflect。"
reflections.jsonl corrupted	Report degraded data state. Process what's readable.
claude-mem MCP tools unavailable	Fall back to keyword matching on JSONL. Note degraded mode in report.
User rejects all proposed diffs	Still mark reflections as distilled. Rejection is data. The report is still valuable as a record.
User wants to modify a diff	Apply the user's override. Record both the proposed value and the user's chosen value.
Gap since last distill is very long (30+ reflections)	Suggest processing in chunks: "你有 [N] 条未处理的复盘记录，建议分批次合成。先处理最近 2 周的？"
Previous distill report has unresolved questions	Carry forward unresolved questions into the new report. Track across reports.
Action experiments have been tried across multiple reflections	Promote successful experiments to criteria or beliefs. Failed experiments → investigate whether the underlying insight was misattributed.
digest_gaps.jsonl has 0 records	Skip cognitive analysis entirely. Report has no "认知盲区分析" chapter.
digest_gaps.jsonl has 1-2 records	Include brief note + "样本不足" caveat. List mastery records if any. Don't generate cognitive_patterns diffs.
digest_gaps.jsonl has 3+ records but all are gaps (no mastery)	Full analysis. "已掌握领域" section: "当前所有校验均有盲区。第一个 mastery 记录会是一个重要的正向锚点。"
digest_gaps.jsonl has records but no re-tests	"持续 gap" section: "暂无 re-test 数据——持续 gap 分析将在首次 digest --retest 后出现。"
digest_gaps.jsonl is corrupt	Report degraded data state. Process what's readable. Skip cognitive analysis if >50% corrupt.
User rejects cognitive_patterns diffs	Don't write cognitive_patterns to DNA. Record rejection in distill report metadata.

Key Files

File	Purpose
references/reflection-protocol.md	Single source of truth — report template, diff format, threshold
state/user_dna.json	Read current model, write accepted diffs
state/reflections.jsonl	Read all reflections, mark as distilled
state/records.jsonl	RAL daily records — context between reflections (note skill)
state/digest_gaps.jsonl	Feynman verification gap reports — cognitive patterns (digest skill, read-only by distill)
state/distill_reports/	Write markdown reports
models/user_dna_schema.py	Value dimension definitions