when-creating-claude-hooks-use-hook-creator

Name: when-creating-claude-hooks-use-hook-creator
Author: DNYoussef

32stars6forks

How to install when-creating-claude-hooks-use-hook-creator?

dnyoussef/context-cascade/when-creating-claude-hooks-use-hook-creator

$npx -y skills add DNYoussef/context-cascade --skill when-creating-claude-hooks-use-hook-creator

Installs into the current project.

Summary

Create Claude Code hooks with proper schemas, RBAC integration, and performance requirements. Use when implementing PreToolUse, PostToolUse, SessionStart, or any of the 10 hook event types for automation, validation, or security enforcement.

Files · 1

View on GitHub

SKILL.md

1	# SKILL: Hook Creator
2
3
4
5	---
6
7	## LIBRARY-FIRST PROTOCOL (MANDATORY)
8
9	Before writing ANY code, you MUST check:
10
11	### Step 1: Library Catalog
12	- Location: `.claude/library/catalog.json`
13	- If match >70%: REUSE or ADAPT
14
15	### Step 2: Patterns Guide
16	- Location: `.claude/docs/inventories/LIBRARY-PATTERNS-GUIDE.md`
17	- If pattern exists: FOLLOW documented approach
18
19	### Step 3: Existing Projects
20	- Location: `D:\Projects\*`
21	- If found: EXTRACT and adapt
22
23	### Decision Matrix
24	\| Match \| Action \|
25	\|-------\|--------\|
26	\| Library >90% \| REUSE directly \|
27	\| Library 70-90% \| ADAPT minimally \|
28	\| Pattern exists \| FOLLOW pattern \|
29	\| In project \| EXTRACT \|
30	\| No match \| BUILD (add to library after) \|
31
32	---
33
34	## Purpose
35
36	Create production-ready Claude Code hooks that integrate with our RBAC security system,
37	follow official schemas, and meet performance requirements (<20ms for pre-hooks).
38
39	## When to Use This Skill
40
41	- Creating new automation hooks (pre/post operations)
42	- Implementing security validation hooks
43	- Building audit/logging hooks
44	- Extending the RBAC permission system
45	- Adding custom session management hooks
46
47	## 8-Stage Hook Creation Methodology
48
49	### Stage 1: Hook Type Selection
50
51	Identify which of the 10 hook event types you need:
52
53	\| Category \| Hook Type \| Purpose \|
54	\|----------\|-----------\|---------\|
55	\| Blocking \| UserPromptSubmit \| Validate/modify user prompts \|
56	\| Blocking \| SessionStart \| Initialize session state \|
57	\| Blocking \| PreToolUse \| Validate tool operations \|
58	\| Blocking \| PermissionRequest \| Auto-approve/deny permissions \|
59	\| Observational \| PostToolUse \| Log tool results \|
60	\| Observational \| Notification \| Forward notifications \|
61	\| Observational \| Stop \| Cleanup on agent stop \|
62	\| Observational \| SubagentStop \| Track subagent completion \|
63	\| Observational \| PreCompact \| Preserve context during compaction \|
64	\| Observational \| SessionEnd \| Final session cleanup \|
65
66	### Stage 2: Schema Definition
67
68	Define input/output schemas based on hook type.
69
70	PreToolUse Input:
71	```json
72	{
73	"session_id": "string",
74	"tool_name": "Bash\|Read\|Write\|Edit\|...",
75	"tool_input": { "...tool-specific..." }
76	}
77	```
78
79	Blocking Output:
80	```json
81	{
82	"continue": true\|false,
83	"decision": "approve\|block\|modify",
84	"reason": "string (if blocked)",
85	"suppressOutput": false,
86	"updatedInput": { "..." }
87	}
88	```
89
90	Non-Blocking Output:
91	```json
92	{
93	"suppressOutput": false
94	}
95	```
96
97	### Stage 3: Template Selection
98
99	Use our pre-built templates:
100
101	- `pre-hook-template.js` - For blocking hooks (PreToolUse, UserPromptSubmit)
102	- `post-hook-template.js` - For observational hooks (PostToolUse, SessionEnd)
103	- `session-hook-template.js` - For session lifecycle hooks
104
105	Generate from templates:
106	```bash
107	node hook-template-generator.js --type pre --name my-validator --event PreToolUse
108	```
109
110	### Stage 4: Core Logic Implementation
111
112	Implement the hook's core logic:
113
114	```javascript
115	#!/usr/bin/env node
116	const fs = require('fs');
117
118	// Read input from stdin
119	const input = JSON.parse(fs.readFileSync(0, 'utf-8'));
120
121	// Your validation/processing logic
122	function processHook(input) {
123	// Implement your logic here
124	return { continue: true, decision: "approve" };
125	}
126
127	// Execute and output result
128	try {
129	const result = processHook(input);
130	console.log(JSON.stringify(result));
131	} catch (error) {
132	console.error(`[HOOK ERROR] ${error.message}`);
133	console.log(JSON.stringify({ continue: true })); // Fail open
134	process.exit(1);
135	}
136	```
137
138	### Stage 5: RBAC Integration
139
140	For security hooks, integrate with our identity system:
141
142	```javascript
143	const { validateAgentIdentity, loadAgentIdentityByName } = require('../utils/identity');
144
145	// Verify agent identity
146	const identity = loadAgentIdentityByName(input.agent_name);
147	const validation = validateAgentIdentity(identity);
148
149	if (!validation.valid) {
150	return {
151	continue: false,
152	decision: "block",
153	reason: `Invalid agent identity: ${validation.errors.join(', ')}`
154	};
155	}
156	```
157
158	### Stage 6: Performance Optimization
159
160	Meet performance targets:
161
162	\| Hook Type \| Target \| Max \|
163	\|-----------\|--------\|-----\|
164	\| Pre-hooks \| <20ms \| 100ms \|
165	\| Post-hooks \| <100ms \| 1000ms \|
166
167	Optimization Patterns:
168	- Cache identity lookups
169	- Avoid synchronous I/O in hot paths
170	- Use matchers to filter events
171	- Batch logging operations
172
173	### Stage 7: Testing
174
175	Create test scenarios:
176
177	```javascript
178	// test-my-hook.js
179	const testCases = [
180	{
181	name: "Should approve valid operation",
182	input: { tool_name: "Read", tool_input: { file_path: "/src/app.js" } },
183	expectedOutput: { continue: true, decision: "approve" }
184	},
185	{
186	name: "Should block dangerous command",
187	input: { tool_name: "B

People who liked this also viewed

6 picks

Type

when-creating-claude-hooks-use-hook-creator

32stars6forks

View on GitHub

How to install when-creating-claude-hooks-use-hook-creator?

dnyoussef/context-cascade/when-creating-claude-hooks-use-hook-creator

$npx -y skills add DNYoussef/context-cascade --skill when-creating-claude-hooks-use-hook-creator

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# SKILL: Hook Creator
2
3
4
5	---
6
7	## LIBRARY-FIRST PROTOCOL (MANDATORY)
8
9	Before writing ANY code, you MUST check:
10
11	### Step 1: Library Catalog
12	- Location: `.claude/library/catalog.json`
13	- If match >70%: REUSE or ADAPT
14
15	### Step 2: Patterns Guide
16	- Location: `.claude/docs/inventories/LIBRARY-PATTERNS-GUIDE.md`
17	- If pattern exists: FOLLOW documented approach
18
19	### Step 3: Existing Projects
20	- Location: `D:\Projects\*`
21	- If found: EXTRACT and adapt
22
23	### Decision Matrix
24	\| Match \| Action \|
25	\|-------\|--------\|
26	\| Library >90% \| REUSE directly \|
27	\| Library 70-90% \| ADAPT minimally \|
28	\| Pattern exists \| FOLLOW pattern \|
29	\| In project \| EXTRACT \|
30	\| No match \| BUILD (add to library after) \|
31
32	---
33
34	## Purpose
35
36	Create production-ready Claude Code hooks that integrate with our RBAC security system,
37	follow official schemas, and meet performance requirements (<20ms for pre-hooks).
38
39	## When to Use This Skill
40
41	- Creating new automation hooks (pre/post operations)
42	- Implementing security validation hooks
43	- Building audit/logging hooks
44	- Extending the RBAC permission system
45	- Adding custom session management hooks
46
47	## 8-Stage Hook Creation Methodology
48
49	### Stage 1: Hook Type Selection
50
51	Identify which of the 10 hook event types you need:
52
53	\| Category \| Hook Type \| Purpose \|
54	\|----------\|-----------\|---------\|
55	\| Blocking \| UserPromptSubmit \| Validate/modify user prompts \|
56	\| Blocking \| SessionStart \| Initialize session state \|
57	\| Blocking \| PreToolUse \| Validate tool operations \|
58	\| Blocking \| PermissionRequest \| Auto-approve/deny permissions \|
59	\| Observational \| PostToolUse \| Log tool results \|
60	\| Observational \| Notification \| Forward notifications \|
61	\| Observational \| Stop \| Cleanup on agent stop \|
62	\| Observational \| SubagentStop \| Track subagent completion \|
63	\| Observational \| PreCompact \| Preserve context during compaction \|
64	\| Observational \| SessionEnd \| Final session cleanup \|
65
66	### Stage 2: Schema Definition
67
68	Define input/output schemas based on hook type.
69
70	PreToolUse Input:
71	```json
72	{
73	"session_id": "string",
74	"tool_name": "Bash\|Read\|Write\|Edit\|...",
75	"tool_input": { "...tool-specific..." }
76	}
77	```
78
79	Blocking Output:
80	```json
81	{
82	"continue": true\|false,
83	"decision": "approve\|block\|modify",
84	"reason": "string (if blocked)",
85	"suppressOutput": false,
86	"updatedInput": { "..." }
87	}
88	```
89
90	Non-Blocking Output:
91	```json
92	{
93	"suppressOutput": false
94	}
95	```
96
97	### Stage 3: Template Selection
98
99	Use our pre-built templates:
100
101	- `pre-hook-template.js` - For blocking hooks (PreToolUse, UserPromptSubmit)
102	- `post-hook-template.js` - For observational hooks (PostToolUse, SessionEnd)
103	- `session-hook-template.js` - For session lifecycle hooks
104
105	Generate from templates:
106	```bash
107	node hook-template-generator.js --type pre --name my-validator --event PreToolUse
108	```
109
110	### Stage 4: Core Logic Implementation
111
112	Implement the hook's core logic:
113
114	```javascript
115	#!/usr/bin/env node
116	const fs = require('fs');
117
118	// Read input from stdin
119	const input = JSON.parse(fs.readFileSync(0, 'utf-8'));
120
121	// Your validation/processing logic
122	function processHook(input) {
123	// Implement your logic here
124	return { continue: true, decision: "approve" };
125	}
126
127	// Execute and output result
128	try {
129	const result = processHook(input);
130	console.log(JSON.stringify(result));
131	} catch (error) {
132	console.error(`[HOOK ERROR] ${error.message}`);
133	console.log(JSON.stringify({ continue: true })); // Fail open
134	process.exit(1);
135	}
136	```
137
138	### Stage 5: RBAC Integration
139
140	For security hooks, integrate with our identity system:
141
142	```javascript
143	const { validateAgentIdentity, loadAgentIdentityByName } = require('../utils/identity');
144
145	// Verify agent identity
146	const identity = loadAgentIdentityByName(input.agent_name);
147	const validation = validateAgentIdentity(identity);
148
149	if (!validation.valid) {
150	return {
151	continue: false,
152	decision: "block",
153	reason: `Invalid agent identity: ${validation.errors.join(', ')}`
154	};
155	}
156	```
157
158	### Stage 6: Performance Optimization
159
160	Meet performance targets:
161
162	\| Hook Type \| Target \| Max \|
163	\|-----------\|--------\|-----\|
164	\| Pre-hooks \| <20ms \| 100ms \|
165	\| Post-hooks \| <100ms \| 1000ms \|
166
167	Optimization Patterns:
168	- Cache identity lookups
169	- Avoid synchronous I/O in hot paths
170	- Use matchers to filter events
171	- Batch logging operations
172
173	### Stage 7: Testing
174
175	Create test scenarios:
176
177	```javascript
178	// test-my-hook.js
179	const testCases = [
180	{
181	name: "Should approve valid operation",
182	input: { tool_name: "Read", tool_input: { file_path: "/src/app.js" } },
183	expectedOutput: { continue: true, decision: "approve" }
184	},
185	{
186	name: "Should block dangerous command",
187	input: { tool_name: "B

People who liked this also viewed

6 picks

Type