prompt-forge

Name: prompt-forge
Author: DNYoussef

32stars6forks

How to install prompt-forge?

dnyoussef/context-cascade/prompt-forge

$npx -y skills add DNYoussef/context-cascade --skill prompt-forge

Installs into the current project.

Summary

/*============================================================================*/ /* PROMPT-FORGE SKILL :: VERILINGUA x VERIX EDITION */ /*============================================================================*/

Files · 1

View on GitHub

SKILL.md

1	/============================================================================/
2	/* PROMPT-FORGE SKILL :: VERILINGUA x VERIX EDITION */
3	/============================================================================/
4
5	---
6	name: prompt-forge
7	version: 2.0.1
8	description: \|
9	[assert\|neutral] Meta-prompt that generates improved prompts and templates. Can improve other prompts including Skill Forge and even itself. All improvements are gated by frozen eval harness. Use when optimizing promp [ground:given] [conf:0.95] [state:confirmed]
10	category: foundry
11	tags:
12	- meta-prompt
13	- self-improvement
14	- recursive
15	- dogfooding
16	- cognitive-frames
17	author: system
18	cognitive_frame:
19	primary: compositional
20	goal_analysis:
21	first_order: "Execute prompt-forge workflow"
22	second_order: "Ensure quality and consistency"
23	third_order: "Enable systematic foundry processes"
24	---
25
26	/----------------------------------------------------------------------------/
27	/* S0 META-IDENTITY */
28	/----------------------------------------------------------------------------/
29
30	[define\|neutral] SKILL := {
31	name: "prompt-forge",
32	category: "foundry",
33	version: "2.0.1",
34	layer: L1
35	} [ground:given] [conf:1.0] [state:confirmed]
36
37	/----------------------------------------------------------------------------/
38	/* S1 COGNITIVE FRAME */
39	/----------------------------------------------------------------------------/
40
41	[define\|neutral] COGNITIVE_FRAME := {
42	frame: "Compositional",
43	source: "German",
44	force: "Build from primitives?"
45	} [ground:cognitive-science] [conf:0.92] [state:confirmed]
46
47	## Kanitsal Cerceve (Evidential Frame Activation)
48	Kaynak dogrulama modu etkin.
49
50	/----------------------------------------------------------------------------/
51	/* S2 TRIGGER CONDITIONS */
52	/----------------------------------------------------------------------------/
53
54	[define\|neutral] TRIGGER_POSITIVE := {
55	keywords: ["prompt-forge", "foundry", "workflow"],
56	context: "user needs prompt-forge capability"
57	} [ground:given] [conf:1.0] [state:confirmed]
58
59	/----------------------------------------------------------------------------/
60	/* S3 CORE CONTENT */
61	/----------------------------------------------------------------------------/
62
63	# Prompt Forge (Meta-Prompt)
64
65	## Kanitsal Cerceve (Evidential Frame Activation)
66	Kaynak dogrulama modu etkin.
67
68
69
70	## Purpose
71
72	Generate improved prompts and templates with:
73	- Explicit rationale for each change
74	- Predicted improvement metrics
75	- Risk assessment
76	- Actionable diffs
77
78	Key Innovation: Can improve Skill Forge prompts, then Skill Forge can improve Prompt Forge prompts - creating a recursive improvement loop.
79
80	## When to Use
81
82	- Optimizing existing prompts for better performance
83	- Creating prompt diffs with clear rationale
84	- Running the recursive improvement loop
85	- Auditing prompts for common issues
86
87	## MCP Requirements
88
89	### memory-mcp (Required)
90
91	Purpose: Store proposals, test results, version history
92
93	Activation:
94	```bash
95	claude mcp add memory-mcp npx @modelcontextprotocol/server-memory
96	```
97
98	---
99
100	## Core Operations
101
102	### Operation 1: Analyze Prompt
103
104	Before improving, deeply understand the target prompt.
105
106	```yaml
107	analysis:
108	target: "{prompt_path}"
109
110	structural_analysis:
111	sections: [list of sections]
112	flow: "How sections connect"
113	dependencies: "What inputs/outputs exist"
114
115	quality_assessment:
116	clarity:
117	score: 0.0-1.0
118	issues: ["Ambiguous instruction in section X"]
119	completeness:
120	score: 0.0-1.0
121	issues: ["Missing failure handling for case Y"]
122	precision:
123	score: 0.0-1.0
124	issues: ["Vague success criteria in section Z"]
125
126	pattern_detection:
127	evidence_based_techniques:
128	self_consistency: present\|missing\|partial
129	program_of_thought: present\|missing\|partial
130	plan_and_solve: present\|missing\|partial
131	failure_handling:
132	explicit_errors: present\|missing\|partial
133	edge_cases: present\|missing\|partial
134	uncertainty: present\|missing\|partial
135
136	improvement_opportunities:
137	- area: "Section X"
138	issue: "Lacks explicit timeout handling"
139	priority: high\|medium\|low
140	predicted_impact: "+X% reliability"
141	```
142
143	### Operation 2: Generate Improvement Proposal
144
145	Create concrete, testable improvement proposals.
146
147	```yaml
148	proposal:
149	id: "prop-{timestamp}"
150	target: "{prompt_path}"
151	type: "prompt_improvement"
152
153	summary: "One-line description of improvement"
154
155	changes:
156	- section: "Section name"
157	location: "Line X-Y"
158	before: \|
159	Original text...
160	after: \|
161	Improved text...
162	rationale: "Why this change improves the prompt"
163	technique: "Which evidence-based technique applied"
164
165	predicted_improvement:
166	primary_metric: "success_rate"
167	expected_delta: "+

People who liked this also viewed

6 picks

Type

prompt-forge

32stars6forks

View on GitHub

How to install prompt-forge?

dnyoussef/context-cascade/prompt-forge

$npx -y skills add DNYoussef/context-cascade --skill prompt-forge

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	/============================================================================/
2	/* PROMPT-FORGE SKILL :: VERILINGUA x VERIX EDITION */
3	/============================================================================/
4
5	---
6	name: prompt-forge
7	version: 2.0.1
8	description: \|
9	[assert\|neutral] Meta-prompt that generates improved prompts and templates. Can improve other prompts including Skill Forge and even itself. All improvements are gated by frozen eval harness. Use when optimizing promp [ground:given] [conf:0.95] [state:confirmed]
10	category: foundry
11	tags:
12	- meta-prompt
13	- self-improvement
14	- recursive
15	- dogfooding
16	- cognitive-frames
17	author: system
18	cognitive_frame:
19	primary: compositional
20	goal_analysis:
21	first_order: "Execute prompt-forge workflow"
22	second_order: "Ensure quality and consistency"
23	third_order: "Enable systematic foundry processes"
24	---
25
26	/----------------------------------------------------------------------------/
27	/* S0 META-IDENTITY */
28	/----------------------------------------------------------------------------/
29
30	[define\|neutral] SKILL := {
31	name: "prompt-forge",
32	category: "foundry",
33	version: "2.0.1",
34	layer: L1
35	} [ground:given] [conf:1.0] [state:confirmed]
36
37	/----------------------------------------------------------------------------/
38	/* S1 COGNITIVE FRAME */
39	/----------------------------------------------------------------------------/
40
41	[define\|neutral] COGNITIVE_FRAME := {
42	frame: "Compositional",
43	source: "German",
44	force: "Build from primitives?"
45	} [ground:cognitive-science] [conf:0.92] [state:confirmed]
46
47	## Kanitsal Cerceve (Evidential Frame Activation)
48	Kaynak dogrulama modu etkin.
49
50	/----------------------------------------------------------------------------/
51	/* S2 TRIGGER CONDITIONS */
52	/----------------------------------------------------------------------------/
53
54	[define\|neutral] TRIGGER_POSITIVE := {
55	keywords: ["prompt-forge", "foundry", "workflow"],
56	context: "user needs prompt-forge capability"
57	} [ground:given] [conf:1.0] [state:confirmed]
58
59	/----------------------------------------------------------------------------/
60	/* S3 CORE CONTENT */
61	/----------------------------------------------------------------------------/
62
63	# Prompt Forge (Meta-Prompt)
64
65	## Kanitsal Cerceve (Evidential Frame Activation)
66	Kaynak dogrulama modu etkin.
67
68
69
70	## Purpose
71
72	Generate improved prompts and templates with:
73	- Explicit rationale for each change
74	- Predicted improvement metrics
75	- Risk assessment
76	- Actionable diffs
77
78	Key Innovation: Can improve Skill Forge prompts, then Skill Forge can improve Prompt Forge prompts - creating a recursive improvement loop.
79
80	## When to Use
81
82	- Optimizing existing prompts for better performance
83	- Creating prompt diffs with clear rationale
84	- Running the recursive improvement loop
85	- Auditing prompts for common issues
86
87	## MCP Requirements
88
89	### memory-mcp (Required)
90
91	Purpose: Store proposals, test results, version history
92
93	Activation:
94	```bash
95	claude mcp add memory-mcp npx @modelcontextprotocol/server-memory
96	```
97
98	---
99
100	## Core Operations
101
102	### Operation 1: Analyze Prompt
103
104	Before improving, deeply understand the target prompt.
105
106	```yaml
107	analysis:
108	target: "{prompt_path}"
109
110	structural_analysis:
111	sections: [list of sections]
112	flow: "How sections connect"
113	dependencies: "What inputs/outputs exist"
114
115	quality_assessment:
116	clarity:
117	score: 0.0-1.0
118	issues: ["Ambiguous instruction in section X"]
119	completeness:
120	score: 0.0-1.0
121	issues: ["Missing failure handling for case Y"]
122	precision:
123	score: 0.0-1.0
124	issues: ["Vague success criteria in section Z"]
125
126	pattern_detection:
127	evidence_based_techniques:
128	self_consistency: present\|missing\|partial
129	program_of_thought: present\|missing\|partial
130	plan_and_solve: present\|missing\|partial
131	failure_handling:
132	explicit_errors: present\|missing\|partial
133	edge_cases: present\|missing\|partial
134	uncertainty: present\|missing\|partial
135
136	improvement_opportunities:
137	- area: "Section X"
138	issue: "Lacks explicit timeout handling"
139	priority: high\|medium\|low
140	predicted_impact: "+X% reliability"
141	```
142
143	### Operation 2: Generate Improvement Proposal
144
145	Create concrete, testable improvement proposals.
146
147	```yaml
148	proposal:
149	id: "prop-{timestamp}"
150	target: "{prompt_path}"
151	type: "prompt_improvement"
152
153	summary: "One-line description of improvement"
154
155	changes:
156	- section: "Section name"
157	location: "Line X-Y"
158	before: \|
159	Original text...
160	after: \|
161	Improved text...
162	rationale: "Why this change improves the prompt"
163	technique: "Which evidence-based technique applied"
164
165	predicted_improvement:
166	primary_metric: "success_rate"
167	expected_delta: "+

People who liked this also viewed

6 picks

Type