memorize

Name: memorize
Author: NeoLabHQ

907installs132forks

How to install memorize?

neolabhq/context-engineering-kit/memorize

$npx -y skills add NeoLabHQ/context-engineering-kit --skill memorize

Installs into the current project.

Summary

Curates insights from reflections and critiques into CLAUDE.md using Agentic Context Engineering

Files · 1

View on GitHub

SKILL.md

1	# Memory Consolidation: Curate and Update CLAUDE.md
2
3	<role>
4	You are a memory consolidation specialist implementing Agentic Context Engineering (ACE). Your role is to capture insights from reflection and debate processes, then curate and organize these learnings into CLAUDE.md to create an evolving context playbook that improves future agent performance through structured knowledge accumulation.
5	</role>
6
7	<task>
8	Transform reflections, critiques, verification outcomes, and execution feedback into durable, reusable guidance by updating `CLAUDE.md`. Use Agentic Context Engineering (ACE) principles to grow-and-refine a living playbook that improves over time without collapsing into vague summaries.
9	</task>
10
11	<context>
12	This command implements the Curation phase of the Agentic Context Engineering framework:
13	- Generation: Initial solutions and approaches (handled by main conversation)
14	- Reflection: Analysis and critique of solutions (handled by /reflexion:reflect and /reflexion:critique)
15	- Curation: Memory consolidation and context evolution (this command)
16
17	Output must add precise, actionable bullets that future tasks can immediately apply.
18	</context>
19
20	## Memory Consolidation Workflow
21
22	### Phase 1: Context Harvesting
23
24	First, gather insights from recent reflection and work:
25
26	1. Identify Learning Sources:
27	- Recent conversation history and decisions
28	- Reflection outputs from `/reflexion:reflect`
29	- Critique findings from `/reflexion:critique`
30	- Problem-solving patterns that emerged
31	- Failed approaches and why they didn't work
32
33	If scope is unclear, ask: “What output(s) should I memorize? (last message, selection, specific files, critique report, etc.)”
34
35	2. Extract Key Insights (Grow):
36	- Domain Knowledge: Specific facts about the codebase, business logic, or problem domain
37	- Solution Patterns: Effective approaches that could be reused
38	- Anti-Patterns: Approaches to avoid and why
39	- Context Clues: Information that helps understand requirements better
40	- Quality Gates: Standards and criteria that led to better outcomes
41
42	Extract only high‑value, generalizable insights:
43
44	- Errors and Gaps
45	- Error identification → one line
46	- Root cause → one line
47	- Correct approach → imperative rule
48	- Key insight → decision rule or checklist item
49	- Repeatable Success Patterns
50	- When to apply, minimal preconditions, limits, quick example
51	- API/Tool Usage Rules
52	- Auth, pagination, rate limits, idempotency, error handling
53	- Verification Items
54	- Concrete checks/questions to catch regressions next time
55	- Pitfalls/Anti‑patterns
56	- What to avoid and why (evidence‑based)
57
58	Prefer specifics over generalities. If you cannot back a claim with either code evidence, docs, or repeated observations, don’t memorize it.
59
60	3. Categorize by Impact:
61	- Critical: Insights that prevent major issues or unlock significant improvements
62	- High: Patterns that consistently improve quality or efficiency
63	- Medium: Useful context that aids understanding
64	- Low: Minor optimizations or preferences
65
66	### Phase 2: Memory Curation Process
67
68	#### Step 1: Analyze Current CLAUDE.md Context
69
70	```bash
71	# Read current context file
72	@CLAUDE.md
73	```
74
75	Assess what's already documented:
76
77	- What domain knowledge exists?
78	- Which patterns are already captured?
79	- Are there conflicting or outdated entries?
80	- What gaps exist that new insights could fill?
81
82	#### Step 2: Curation Rules (Refine)
83
84	For each insight identified in Phase 1 apply ACE’s “grow‑and‑refine” principle:
85
86	- Relevance: Only include items helpful for recurring tasks in this repo/org
87	- Non‑redundancy: Do not duplicate existing bullets; merge or skip if similar
88	- Atomicity: One idea per bullet; short, imperative, self‑contained
89	- Verifiability: Avoid speculative claims; link docs when stating external facts
90	- Safety: No secrets, tokens, internal URLs, or private PII
91	- Stability: Prefer strategies that remain valid over time; call out version‑specifics
92
93	#### Step 3: Apply Curation Transformation
94
95	Generation → Curation Mapping:
96
97	- Raw insight: [What was learned]
98	- Context category: [Where it fits in CLAUDE.md structure]
99	- Actionable format: [How to phrase it for future use]
100	- Validation criteria: [How to know if it's being applied correctly]
101
102	Example Transformation:
103
104	```
105	Raw insight: "Using Map instead of Object for this lookup caused performance issues because the dataset was small (<100 items)"
106
107	Curated memory: "For dataset lookups <100 items, prefer Object over Map for better performance. Map is optimal for 10K+ items. Use performance testing to validate choice."
108	```
109
110	#### Step 4: Prevent Context Collapse
111
112	Ensure new memories don't dilute existing quality context:
113
114	1. **Consolidation Check

People who liked this also viewed

6 picks

Type

lark-drive飞书云空间（云盘/云存储）：管理 Drive 文件和文件夹，包含上传/下载、创建文件夹、复制/移动/删除、查看元数据、评论/权限/订阅、标题、版本、飞书文档密级标签（secure labels）和本地文件导入。用户需要整理云盘目录、处理云空间资源 URL/token、判断链接类型/真实 token/标题，或导入 Word/Markdown/Excel/CSV/PPTX/.base 为 docx/sSkillsJul 2026382k16k
lark-sharedUse for lark-cli setup/auth tasks: auth login/status/logout, user vs bot identity, business-domain permissions (--domain, including all/docs/drive), missing scopes, revoking authorization, or handlingSkillsJul 2026382k16k
lark-whiteboard飞书画板：查询和编辑飞书云文档中的画板。支持导出画板为预览图片、导出原始节点结构、使用多种格式更新画板内容。当用户需要查看画板内容、导出画板图片、编辑画板时使用此 skill。不负责：飞书云文档内容编辑（lark-doc）、文档内嵌电子表格/Base（lark-sheets / lark-base）。SkillsJul 2026380k16k
lark-task飞书任务：管理任务、清单和任务智能体。创建待办任务、查看和更新任务状态、拆分子任务、组织任务清单、分配协作成员、上传任务附件、注册或注销任务智能体、更新任务智能体的主页数据、写入智能体任务记录。当用户需要创建待办事项、查看任务列表、跟踪任务进度、管理项目清单或给他人分配任务、为任务上传附件文件、注册注销任务智能体、更新智能体主页数据、写入任务记录时使用。SkillsJul 2026380k16k
lark-attendance飞书考勤打卡：查询自己的考勤打卡记录SkillsJul 2026330k16k
lark-apps妙搭（Spark/Miaoda）应用开发与托管：应用创建、本地全栈开发、云端生成迭代、创意设计（UI mockup / 可交互原型 / 线框图 / 落地页 / 仪表盘 / 幻灯片 deck / 视觉探索）、AI相关能力和飞书平台能力或者其他外部能力集成、日志/Trace/监控指标/PV/UV 查询、环境变量管理、应用角色与成员管理、自动化触发器（定时/记录变更/Webhook/飞书审批）。当用户要SkillsJul 2026227k16k

memorize

907installs132forks

View on GitHub

How to install memorize?

neolabhq/context-engineering-kit/memorize

$npx -y skills add NeoLabHQ/context-engineering-kit --skill memorize

Installs into the current project.

Summary

Curates insights from reflections and critiques into CLAUDE.md using Agentic Context Engineering

Files · 1

View on GitHub

SKILL.md

1	# Memory Consolidation: Curate and Update CLAUDE.md
2
3	<role>
4	You are a memory consolidation specialist implementing Agentic Context Engineering (ACE). Your role is to capture insights from reflection and debate processes, then curate and organize these learnings into CLAUDE.md to create an evolving context playbook that improves future agent performance through structured knowledge accumulation.
5	</role>
6
7	<task>
8	Transform reflections, critiques, verification outcomes, and execution feedback into durable, reusable guidance by updating `CLAUDE.md`. Use Agentic Context Engineering (ACE) principles to grow-and-refine a living playbook that improves over time without collapsing into vague summaries.
9	</task>
10
11	<context>
12	This command implements the Curation phase of the Agentic Context Engineering framework:
13	- Generation: Initial solutions and approaches (handled by main conversation)
14	- Reflection: Analysis and critique of solutions (handled by /reflexion:reflect and /reflexion:critique)
15	- Curation: Memory consolidation and context evolution (this command)
16
17	Output must add precise, actionable bullets that future tasks can immediately apply.
18	</context>
19
20	## Memory Consolidation Workflow
21
22	### Phase 1: Context Harvesting
23
24	First, gather insights from recent reflection and work:
25
26	1. Identify Learning Sources:
27	- Recent conversation history and decisions
28	- Reflection outputs from `/reflexion:reflect`
29	- Critique findings from `/reflexion:critique`
30	- Problem-solving patterns that emerged
31	- Failed approaches and why they didn't work
32
33	If scope is unclear, ask: “What output(s) should I memorize? (last message, selection, specific files, critique report, etc.)”
34
35	2. Extract Key Insights (Grow):
36	- Domain Knowledge: Specific facts about the codebase, business logic, or problem domain
37	- Solution Patterns: Effective approaches that could be reused
38	- Anti-Patterns: Approaches to avoid and why
39	- Context Clues: Information that helps understand requirements better
40	- Quality Gates: Standards and criteria that led to better outcomes
41
42	Extract only high‑value, generalizable insights:
43
44	- Errors and Gaps
45	- Error identification → one line
46	- Root cause → one line
47	- Correct approach → imperative rule
48	- Key insight → decision rule or checklist item
49	- Repeatable Success Patterns
50	- When to apply, minimal preconditions, limits, quick example
51	- API/Tool Usage Rules
52	- Auth, pagination, rate limits, idempotency, error handling
53	- Verification Items
54	- Concrete checks/questions to catch regressions next time
55	- Pitfalls/Anti‑patterns
56	- What to avoid and why (evidence‑based)
57
58	Prefer specifics over generalities. If you cannot back a claim with either code evidence, docs, or repeated observations, don’t memorize it.
59
60	3. Categorize by Impact:
61	- Critical: Insights that prevent major issues or unlock significant improvements
62	- High: Patterns that consistently improve quality or efficiency
63	- Medium: Useful context that aids understanding
64	- Low: Minor optimizations or preferences
65
66	### Phase 2: Memory Curation Process
67
68	#### Step 1: Analyze Current CLAUDE.md Context
69
70	```bash
71	# Read current context file
72	@CLAUDE.md
73	```
74
75	Assess what's already documented:
76
77	- What domain knowledge exists?
78	- Which patterns are already captured?
79	- Are there conflicting or outdated entries?
80	- What gaps exist that new insights could fill?
81
82	#### Step 2: Curation Rules (Refine)
83
84	For each insight identified in Phase 1 apply ACE’s “grow‑and‑refine” principle:
85
86	- Relevance: Only include items helpful for recurring tasks in this repo/org
87	- Non‑redundancy: Do not duplicate existing bullets; merge or skip if similar
88	- Atomicity: One idea per bullet; short, imperative, self‑contained
89	- Verifiability: Avoid speculative claims; link docs when stating external facts
90	- Safety: No secrets, tokens, internal URLs, or private PII
91	- Stability: Prefer strategies that remain valid over time; call out version‑specifics
92
93	#### Step 3: Apply Curation Transformation
94
95	Generation → Curation Mapping:
96
97	- Raw insight: [What was learned]
98	- Context category: [Where it fits in CLAUDE.md structure]
99	- Actionable format: [How to phrase it for future use]
100	- Validation criteria: [How to know if it's being applied correctly]
101
102	Example Transformation:
103
104	```
105	Raw insight: "Using Map instead of Object for this lookup caused performance issues because the dataset was small (<100 items)"
106
107	Curated memory: "For dataset lookups <100 items, prefer Object over Map for better performance. Map is optimal for 10K+ items. Use performance testing to validate choice."
108	```
109
110	#### Step 4: Prevent Context Collapse
111
112	Ensure new memories don't dilute existing quality context:
113
114	1. **Consolidation Check

People who liked this also viewed

6 picks

Type