propose-hypotheses

Name: propose-hypotheses
Author: NeoLabHQ

826installs132forks

How to install propose-hypotheses?

neolabhq/context-engineering-kit/propose-hypotheses

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

Installs into the current project.

Summary

Execute complete FPF cycle from hypothesis generation to decision

Files · 1

View on GitHub

SKILL.md

1	# Propose Hypotheses Workflow
2
3	Execute the First Principles Framework (FPF) cycle: generate competing hypotheses, verify logic, validate evidence, audit trust, and produce a decision.
4
5	## User Input
6
7	```text
8	Problem Statement: $ARGUMENTS
9	```
10
11	## Workflow Execution
12
13	### Step 1a: Create Directory Structure (Main Agent)
14
15	Create `.fpf/` directory structure if it does not exist:
16
17	```bash
18	mkdir -p .fpf/{evidence,decisions,sessions,knowledge/{L0,L1,L2,invalid}}
19	touch .fpf/{evidence,decisions,sessions,knowledge/{L0,L1,L2,invalid}}/.gitkeep
20	```
21
22	Postcondition: `.fpf/` directory scaffold exists.
23
24	---
25
26	### Step 1b: Initialize Context (FPF Agent)
27
28	Launch fpf-agent with sonnet[1m] model:
29	- Description: "Initialize FPF context"
30	- Prompt:
31	```
32	Read ${CLAUDE_PLUGIN_ROOT}/tasks/init-context.md and execute.
33
34	Problem Statement: $ARGUMENTS
35
36	Write: Context summary to `.fpf/context.md`**
37	```
38
39	---
40
41	### Step 2: Generate Hypotheses (FPF Agent)
42
43	Launch fpf-agent with sonnet[1m] model:
44	- Description: "Generate L0 hypotheses"
45	- Prompt:
46	```
47	Read ${CLAUDE_PLUGIN_ROOT}/tasks/generate-hypotheses.md and execute.
48
49	Problem Statement: $ARGUMENTS
50	Context: <summary from Step 1b>
51
52	Write: List of hypothesis IDs and titles to `.fpf/knowledge/L0/`
53
54	Reply with summary table in markdown format:
55
56	\| ID \| Title \| Kind \| Scope \|
57	\|----\|-------\|------\|-------\|
58	\| ... \| ... \| ... \| ... \|
59	```
60
61	---
62
63	### Step 3: Present Summary (Main Agent)
64
65	1. Read all L0 hypothesis files from `.fpf/knowledge/L0/`
66	2. Present summary table from agent response.
67	3. Ask user: "Would you like to add any hypotheses of your own? (yes/no)"
68
69	---
70
71	### Step 4: Add User Hypothesis (FPF Agent, Conditional Loop)
72
73	Condition: User says yes to adding hypotheses.
74
75	Launch fpf-agent with sonnet[1m] model:
76	- Description: "Add user hypothesis"
77	- Prompt:
78	```
79	Read ${CLAUDE_PLUGIN_ROOT}/tasks/add-user-hypothesis.md and execute.
80
81	User Hypothesis Description: <get from user>
82
83	Write: User hypothesis to `.fpf/knowledge/L0/`
84	```
85
86	Loop: Return to Step 3 after hypothesis is added.
87
88	Exit: When user says no or declines to add more.
89
90	---
91
92	### Step 5: Verify Logic (Parallel Sub-Agents)
93
94	Condition: User finished adding hypotheses.
95
96	For EACH L0 hypothesis file in `.fpf/knowledge/L0/`, launch parallel fpf-agent with sonnet[1m] model:
97	- Description: "Verify hypothesis: <hypothesis-id>"
98	- Prompt:
99	```
100	Read ${CLAUDE_PLUGIN_ROOT}/tasks/verify-logic.md and execute.
101
102	Hypothesis ID: <hypothesis-id>
103	Hypothesis File: .fpf/knowledge/L0/<hypothesis-id>.md
104
105	Move: After you complete verification, move the file to `.fpf/knowledge/L1/` or `.fpf/knowledge/invalid/`.
106	```
107
108	Wait for all agents, then check that files are moved to `.fpf/knowledge/L1/` or `.fpf/knowledge/invalid/`.
109
110	---
111
112	### Step 6: Validate Evidence (Parallel Sub-Agents)
113
114	For EACH L1 hypothesis file in `.fpf/knowledge/L1/`, launch parallel fpf-agent with sonnet[1m] model:
115	- Description: "Validate hypothesis: <hypothesis-id>"
116	- Prompt:
117	```
118	Read ${CLAUDE_PLUGIN_ROOT}/tasks/validate-evidence.md and execute.
119
120	Hypothesis ID: <hypothesis-id>
121	Hypothesis File: .fpf/knowledge/L1/<hypothesis-id>.md
122
123	Move: After you complete validation, move the file to `.fpf/knowledge/L2/` or `.fpf/knowledge/invalid/`.
124	```
125
126	Wait for all agents, then check that files are moved to `.fpf/knowledge/L2/` or `.fpf/knowledge/invalid/`.
127
128	---
129
130	### Step 7: Audit Trust (Parallel Sub-Agents)
131
132	For EACH L2 hypothesis file in `.fpf/knowledge/L2/`, launch parallel fpf-agent with sonnet[1m] model:
133	- Description: "Audit trust: <hypothesis-id>"
134	- Prompt:
135	```
136	Read ${CLAUDE_PLUGIN_ROOT}/tasks/audit-trust.md and execute.
137
138	Hypothesis ID: <hypothesis-id>
139	Hypothesis File: .fpf/knowledge/L2/<hypothesis-id>.md
140
141	Write: Audit report to `.fpf/evidence/audit-{hypothesis-id}-{YYYY-MM-DD}.md`
142
143	Reply: with R_eff score and weakest link
144	```
145
146	Wait for all agents, then check that audit reports are created in `.fpf/evidence/`.
147
148	---
149
150	### Step 8: Make Decision (FPF Agent)
151
152	Launch fpf-agent with sonnet[1m] model:
153	- Description: "Create decision record"
154	- Prompt:
155	```
156	Read ${CLAUDE_PLUGIN_ROOT}/tasks/decide.md and execute.
157
158	Problem Statement: $ARGUMENTS
159	L2 Hypotheses Directory: .fpf/knowledge/L2/
160	Audit Reports: .fpf/evidence/
161
162	Write: Decision record to `.fpf/decisions/`
163
164	Reply: with decision record summary in markdown format:
165
166	\| Hypothesis \| R_eff \| Weakest Link \| Status \|
167	\|------------\|-------\|--------------\|--------\|
168	\| ... \| ... \| ... \| ... \|
169
170	Recommended Decision: <hypothesis title>
171
172	Rationale: <brief explanation>
173	```
174
175	Wait for agent, then check that decision record is created in `.fpf/decision

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

propose-hypotheses

826installs132forks

View on GitHub

How to install propose-hypotheses?

neolabhq/context-engineering-kit/propose-hypotheses

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

Installs into the current project.

Summary

Execute complete FPF cycle from hypothesis generation to decision

Files · 1

View on GitHub

SKILL.md

1	# Propose Hypotheses Workflow
2
3	Execute the First Principles Framework (FPF) cycle: generate competing hypotheses, verify logic, validate evidence, audit trust, and produce a decision.
4
5	## User Input
6
7	```text
8	Problem Statement: $ARGUMENTS
9	```
10
11	## Workflow Execution
12
13	### Step 1a: Create Directory Structure (Main Agent)
14
15	Create `.fpf/` directory structure if it does not exist:
16
17	```bash
18	mkdir -p .fpf/{evidence,decisions,sessions,knowledge/{L0,L1,L2,invalid}}
19	touch .fpf/{evidence,decisions,sessions,knowledge/{L0,L1,L2,invalid}}/.gitkeep
20	```
21
22	Postcondition: `.fpf/` directory scaffold exists.
23
24	---
25
26	### Step 1b: Initialize Context (FPF Agent)
27
28	Launch fpf-agent with sonnet[1m] model:
29	- Description: "Initialize FPF context"
30	- Prompt:
31	```
32	Read ${CLAUDE_PLUGIN_ROOT}/tasks/init-context.md and execute.
33
34	Problem Statement: $ARGUMENTS
35
36	Write: Context summary to `.fpf/context.md`**
37	```
38
39	---
40
41	### Step 2: Generate Hypotheses (FPF Agent)
42
43	Launch fpf-agent with sonnet[1m] model:
44	- Description: "Generate L0 hypotheses"
45	- Prompt:
46	```
47	Read ${CLAUDE_PLUGIN_ROOT}/tasks/generate-hypotheses.md and execute.
48
49	Problem Statement: $ARGUMENTS
50	Context: <summary from Step 1b>
51
52	Write: List of hypothesis IDs and titles to `.fpf/knowledge/L0/`
53
54	Reply with summary table in markdown format:
55
56	\| ID \| Title \| Kind \| Scope \|
57	\|----\|-------\|------\|-------\|
58	\| ... \| ... \| ... \| ... \|
59	```
60
61	---
62
63	### Step 3: Present Summary (Main Agent)
64
65	1. Read all L0 hypothesis files from `.fpf/knowledge/L0/`
66	2. Present summary table from agent response.
67	3. Ask user: "Would you like to add any hypotheses of your own? (yes/no)"
68
69	---
70
71	### Step 4: Add User Hypothesis (FPF Agent, Conditional Loop)
72
73	Condition: User says yes to adding hypotheses.
74
75	Launch fpf-agent with sonnet[1m] model:
76	- Description: "Add user hypothesis"
77	- Prompt:
78	```
79	Read ${CLAUDE_PLUGIN_ROOT}/tasks/add-user-hypothesis.md and execute.
80
81	User Hypothesis Description: <get from user>
82
83	Write: User hypothesis to `.fpf/knowledge/L0/`
84	```
85
86	Loop: Return to Step 3 after hypothesis is added.
87
88	Exit: When user says no or declines to add more.
89
90	---
91
92	### Step 5: Verify Logic (Parallel Sub-Agents)
93
94	Condition: User finished adding hypotheses.
95
96	For EACH L0 hypothesis file in `.fpf/knowledge/L0/`, launch parallel fpf-agent with sonnet[1m] model:
97	- Description: "Verify hypothesis: <hypothesis-id>"
98	- Prompt:
99	```
100	Read ${CLAUDE_PLUGIN_ROOT}/tasks/verify-logic.md and execute.
101
102	Hypothesis ID: <hypothesis-id>
103	Hypothesis File: .fpf/knowledge/L0/<hypothesis-id>.md
104
105	Move: After you complete verification, move the file to `.fpf/knowledge/L1/` or `.fpf/knowledge/invalid/`.
106	```
107
108	Wait for all agents, then check that files are moved to `.fpf/knowledge/L1/` or `.fpf/knowledge/invalid/`.
109
110	---
111
112	### Step 6: Validate Evidence (Parallel Sub-Agents)
113
114	For EACH L1 hypothesis file in `.fpf/knowledge/L1/`, launch parallel fpf-agent with sonnet[1m] model:
115	- Description: "Validate hypothesis: <hypothesis-id>"
116	- Prompt:
117	```
118	Read ${CLAUDE_PLUGIN_ROOT}/tasks/validate-evidence.md and execute.
119
120	Hypothesis ID: <hypothesis-id>
121	Hypothesis File: .fpf/knowledge/L1/<hypothesis-id>.md
122
123	Move: After you complete validation, move the file to `.fpf/knowledge/L2/` or `.fpf/knowledge/invalid/`.
124	```
125
126	Wait for all agents, then check that files are moved to `.fpf/knowledge/L2/` or `.fpf/knowledge/invalid/`.
127
128	---
129
130	### Step 7: Audit Trust (Parallel Sub-Agents)
131
132	For EACH L2 hypothesis file in `.fpf/knowledge/L2/`, launch parallel fpf-agent with sonnet[1m] model:
133	- Description: "Audit trust: <hypothesis-id>"
134	- Prompt:
135	```
136	Read ${CLAUDE_PLUGIN_ROOT}/tasks/audit-trust.md and execute.
137
138	Hypothesis ID: <hypothesis-id>
139	Hypothesis File: .fpf/knowledge/L2/<hypothesis-id>.md
140
141	Write: Audit report to `.fpf/evidence/audit-{hypothesis-id}-{YYYY-MM-DD}.md`
142
143	Reply: with R_eff score and weakest link
144	```
145
146	Wait for all agents, then check that audit reports are created in `.fpf/evidence/`.
147
148	---
149
150	### Step 8: Make Decision (FPF Agent)
151
152	Launch fpf-agent with sonnet[1m] model:
153	- Description: "Create decision record"
154	- Prompt:
155	```
156	Read ${CLAUDE_PLUGIN_ROOT}/tasks/decide.md and execute.
157
158	Problem Statement: $ARGUMENTS
159	L2 Hypotheses Directory: .fpf/knowledge/L2/
160	Audit Reports: .fpf/evidence/
161
162	Write: Decision record to `.fpf/decisions/`
163
164	Reply: with decision record summary in markdown format:
165
166	\| Hypothesis \| R_eff \| Weakest Link \| Status \|
167	\|------------\|-------\|--------------\|--------\|
168	\| ... \| ... \| ... \| ... \|
169
170	Recommended Decision: <hypothesis title>
171
172	Rationale: <brief explanation>
173	```
174
175	Wait for agent, then check that decision record is created in `.fpf/decision

People who liked this also viewed

6 picks

Type