sylph-create-skill

Name: sylph-create-skill
Author: getnao

168stars52forks

How to install sylph-create-skill?

getnao/sylph/sylph-create-skill

$npx -y skills add getnao/sylph --skill sylph-create-skill

Installs into the current project.

Summary

Create a new skill from scratch. Generates the SKILL.md, registers it in AGENTS.md, and sets up the content folder structure.

Files · 1

View on GitHub

SKILL.md

1	# Create Skill
2
3	## When to use
4
5	Run `/sylph-create-skill <name>` when the CAO wants to add a new skill that doesn't exist yet.
6
7	## Usage
8
9	```
10	/sylph-create-skill social-listening
11	/sylph-create-skill competitor-analysis
12	```
13
14	---
15
16	## Step 1: Understand the skill
17
18	Ask the CAO ONE question:
19
20	> What should [skill-name] do? (2-3 sentences is enough)
21
22	From her answer, derive:
23	- Purpose: what problem does it solve?
24	- Domain: which area does it belong to? (content, sales, product, brand, finance, hr, legal, ops, events, writing, agents)
25	- Trigger: when would someone invoke it? (on-demand, daily, event-triggered)
26	- Inputs: what does it need to run? (a topic, a file, a customer name, nothing)
27	- Outputs: what does it produce? (a draft, a report, an issue, a Slack message)
28	- MCP connectors: which external tools does it need?
29
30	## Step 2: Create the skill file
31
32	Create `.claude/skills/<skill-name>/SKILL.md` with this structure:
33
34	```markdown
35	---
36	name: <skill-name>
37	description: <one line - what it does and when to use it>
38	---
39
40	# <Skill Title>
41
42	## MCP connectors
43
44	\| Connector \| Purpose \|
45	\|-----------\|---------\|
46	\| [tool] \| [what it's used for] \|
47
48	## Context loading
49
50	1. Read [relevant files] for context
51	2. Read `_insights.md` for the domain (if exists)
52	3. Read `_examples/` for quality reference (if exists)
53
54	## Arguments
55
56	\| Argument \| What it does \|
57	\|----------\|-------------\|
58	\| [arg] \| [description] \|
59
60	## Execution
61
62	1. [Step 1]
63	2. [Step 2]
64	3. [Step 3]
65
66	## Guardrails
67
68	- [Safety rule 1]
69	- [Safety rule 2]
70	- Never publish or send directly - always save to `_drafts/`
71
72	## Self-improvement
73
74	After the CAO reviews and approves the output:
75
76	1. Move the final version from `_drafts/` to `_published/`
77	2. Diff what was drafted vs what the CAO kept - identify patterns in her edits
78	3. Update `_insights.md` with what worked, what she changed, and why
79	4. If her edits reveal a recurring pattern or new rule, add it to this skill file
80	5. If the output was a top performer, promote it to `_examples/`
81	```
82
83	### Skill file rules
84
85	- Flat path: `.claude/skills/<skill-name>/SKILL.md` - one level only, no subdirectories
86	- YAML frontmatter required: `name` and `description` fields
87	- MCP connectors table: always include, even if empty ("None required")
88	- Guardrails section: always include - every skill must define what it does NOT do
89	- No em dashes: use `-` or `:`
90
91	## Step 3: Create the content folder (if applicable)
92
93	If the skill produces content, create the folder structure:
94
95	```
96	<domain>/<channel>/
97	_drafts/ # where new content goes
98	_published/ # approved content archive
99	_examples/ # best-performing pieces (few-shot reference)
100	_insights.md # channel learnings (create empty with header)
101	```
102
103	Not all skills need a content folder. Agent skills, ops skills, and tool skills may just use `agents/<name>/_logs/`.
104
105	## Step 4: Register the skill
106
107	Add the skill to the index table in `AGENTS.md` (section 8):
108
109	```markdown
110	\| <skill-name> \| <domain> \| `.claude/skills/<skill-name>/SKILL.md` \|
111	```
112
113	Insert in alphabetical order within the domain group.
114
115	## Step 5: Create the setup guide (optional)
116
117	If the skill needs configuration (MCP connections, examples, voice derivation), create a setup guide at `.claude/skills/sylph-setup-skill/setup-<skill-name>.md` and add it to the dispatch table in `sylph-setup-skill/SKILL.md`.
118
119	Only create a setup guide if the skill requires initial configuration. Simple on-demand skills don't need one.
120
121	## Step 6: Add the self-improvement loop (mandatory)
122
123	Every skill MUST have a `## Self-improvement` section as its last section. This is what makes Sylph a self-improving system instead of a static prompt library.
124
125	Add this section to the skill file you created in Step 2:
126
127	```markdown
128	## Self-improvement
129
130	After the CAO reviews and approves the output:
131
132	1. Move the final version from `_drafts/` to `_published/`
133	2. Diff what was drafted vs what the CAO kept - identify patterns in her edits
134	3. Update the relevant `_insights.md` with what worked, what she changed, and why
135	4. If her edits reveal a recurring pattern or new rule, add it to this skill file
136	5. If the output was a top performer, promote it to `_examples/` with annotated frontmatter
137	```
138
139	Adapt the wording to fit the skill's output type:
140	- Content skills (posts, emails, articles): focus on voice, structure, and tone patterns
141	- Agent/routine skills (CoS, CSM, EA): focus on format, scope, and prioritization patterns
142	- Analysis skills (reports, screening, reviews): focus on depth, framing, and what the CAO found useful vs noise
143	- Ops skills (CRM, docs updates): focus on process efficiency and what the CAO corrected
144
145	The key principle: the skill improves because the CAO's edits are signal. Capture what she changed and why, then

People who liked this also viewed

6 picks

Type

paperclip-create-agentCreate new agents in Paperclip with governance-aware hiring. Use when you need to inspect adapter configuration options, compare existing agent configs, draft a new agent prompt/config, and submit a hSkillsJul 2026157k5
subagent-driven-developmentUse when executing implementation plans with independent tasks in the current sessionSkillsJul 2026154k259k
dispatching-parallel-agentsUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesSkillsJul 2026143k259k
browser-act-skill-forgeForges reusable Skill packages (SKILL.md + scripts) from website exploration via browser-act — no re-exploration later. Use when: user wants a reusable Skill for any website, needs to understand a sitSkillsJul 202685k4.6k
wizardGenerate an interactive bash wizard that walks a human through a manual procedure — third-party setup, a one-off migration, an A→B state transition — opening URLs, capturing values, confirming each stSkillsJul 202675k182k
wonda-cliUsing the Wonda CLI to generate images, videos, music, and audio from the terminal — plus LinkedIn, Reddit, and X/Twitter research and automationSkillsJul 202666k137

sylph-create-skill

168stars52forks

View on GitHub

How to install sylph-create-skill?

getnao/sylph/sylph-create-skill

$npx -y skills add getnao/sylph --skill sylph-create-skill

Installs into the current project.

Summary

Create a new skill from scratch. Generates the SKILL.md, registers it in AGENTS.md, and sets up the content folder structure.

Files · 1

View on GitHub

SKILL.md

1	# Create Skill
2
3	## When to use
4
5	Run `/sylph-create-skill <name>` when the CAO wants to add a new skill that doesn't exist yet.
6
7	## Usage
8
9	```
10	/sylph-create-skill social-listening
11	/sylph-create-skill competitor-analysis
12	```
13
14	---
15
16	## Step 1: Understand the skill
17
18	Ask the CAO ONE question:
19
20	> What should [skill-name] do? (2-3 sentences is enough)
21
22	From her answer, derive:
23	- Purpose: what problem does it solve?
24	- Domain: which area does it belong to? (content, sales, product, brand, finance, hr, legal, ops, events, writing, agents)
25	- Trigger: when would someone invoke it? (on-demand, daily, event-triggered)
26	- Inputs: what does it need to run? (a topic, a file, a customer name, nothing)
27	- Outputs: what does it produce? (a draft, a report, an issue, a Slack message)
28	- MCP connectors: which external tools does it need?
29
30	## Step 2: Create the skill file
31
32	Create `.claude/skills/<skill-name>/SKILL.md` with this structure:
33
34	```markdown
35	---
36	name: <skill-name>
37	description: <one line - what it does and when to use it>
38	---
39
40	# <Skill Title>
41
42	## MCP connectors
43
44	\| Connector \| Purpose \|
45	\|-----------\|---------\|
46	\| [tool] \| [what it's used for] \|
47
48	## Context loading
49
50	1. Read [relevant files] for context
51	2. Read `_insights.md` for the domain (if exists)
52	3. Read `_examples/` for quality reference (if exists)
53
54	## Arguments
55
56	\| Argument \| What it does \|
57	\|----------\|-------------\|
58	\| [arg] \| [description] \|
59
60	## Execution
61
62	1. [Step 1]
63	2. [Step 2]
64	3. [Step 3]
65
66	## Guardrails
67
68	- [Safety rule 1]
69	- [Safety rule 2]
70	- Never publish or send directly - always save to `_drafts/`
71
72	## Self-improvement
73
74	After the CAO reviews and approves the output:
75
76	1. Move the final version from `_drafts/` to `_published/`
77	2. Diff what was drafted vs what the CAO kept - identify patterns in her edits
78	3. Update `_insights.md` with what worked, what she changed, and why
79	4. If her edits reveal a recurring pattern or new rule, add it to this skill file
80	5. If the output was a top performer, promote it to `_examples/`
81	```
82
83	### Skill file rules
84
85	- Flat path: `.claude/skills/<skill-name>/SKILL.md` - one level only, no subdirectories
86	- YAML frontmatter required: `name` and `description` fields
87	- MCP connectors table: always include, even if empty ("None required")
88	- Guardrails section: always include - every skill must define what it does NOT do
89	- No em dashes: use `-` or `:`
90
91	## Step 3: Create the content folder (if applicable)
92
93	If the skill produces content, create the folder structure:
94
95	```
96	<domain>/<channel>/
97	_drafts/ # where new content goes
98	_published/ # approved content archive
99	_examples/ # best-performing pieces (few-shot reference)
100	_insights.md # channel learnings (create empty with header)
101	```
102
103	Not all skills need a content folder. Agent skills, ops skills, and tool skills may just use `agents/<name>/_logs/`.
104
105	## Step 4: Register the skill
106
107	Add the skill to the index table in `AGENTS.md` (section 8):
108
109	```markdown
110	\| <skill-name> \| <domain> \| `.claude/skills/<skill-name>/SKILL.md` \|
111	```
112
113	Insert in alphabetical order within the domain group.
114
115	## Step 5: Create the setup guide (optional)
116
117	If the skill needs configuration (MCP connections, examples, voice derivation), create a setup guide at `.claude/skills/sylph-setup-skill/setup-<skill-name>.md` and add it to the dispatch table in `sylph-setup-skill/SKILL.md`.
118
119	Only create a setup guide if the skill requires initial configuration. Simple on-demand skills don't need one.
120
121	## Step 6: Add the self-improvement loop (mandatory)
122
123	Every skill MUST have a `## Self-improvement` section as its last section. This is what makes Sylph a self-improving system instead of a static prompt library.
124
125	Add this section to the skill file you created in Step 2:
126
127	```markdown
128	## Self-improvement
129
130	After the CAO reviews and approves the output:
131
132	1. Move the final version from `_drafts/` to `_published/`
133	2. Diff what was drafted vs what the CAO kept - identify patterns in her edits
134	3. Update the relevant `_insights.md` with what worked, what she changed, and why
135	4. If her edits reveal a recurring pattern or new rule, add it to this skill file
136	5. If the output was a top performer, promote it to `_examples/` with annotated frontmatter
137	```
138
139	Adapt the wording to fit the skill's output type:
140	- Content skills (posts, emails, articles): focus on voice, structure, and tone patterns
141	- Agent/routine skills (CoS, CSM, EA): focus on format, scope, and prioritization patterns
142	- Analysis skills (reports, screening, reviews): focus on depth, framing, and what the CAO found useful vs noise
143	- Ops skills (CRM, docs updates): focus on process efficiency and what the CAO corrected
144
145	The key principle: the skill improves because the CAO's edits are signal. Capture what she changed and why, then

People who liked this also viewed

6 picks

Type