skill-builder

Name: skill-builder
Author: testdouble

122stars12forks

How to install skill-builder?

testdouble/han/skill-builder

$npx -y skills add testdouble/han --skill skill-builder

Installs into the current project.

Summary

Builds a new Claude Code skill from scratch through a relentless, evidence-based interview that walks the skill's design tree decision-by-decision — entity fit, use cases, name, description, workflow steps, tools, and progressive-disclosure layout — then reviews the finished skil

Files · 1

View on GitHub

SKILL.md

1	## Guidance Location
2
3	The authoritative skill-authoring guidance ships in this plugin. Read the
4	specific document a decision needs, when that decision is on the table — never
5	read them all up front, because that defeats progressive disclosure and burns
6	context on guidance the current skill does not touch.
7
8	- Plugin-building guidance root: `${CLAUDE_PLUGIN_ROOT}/skills/guidance/references/`
9	- Skill-specific guidance: `${CLAUDE_PLUGIN_ROOT}/skills/guidance/references/skill-building-guidance/`
10
11	Map from decision to governing document (read just-in-time):
12
13	\| Decision on the table \| Read \|
14	\|---\|---\|
15	\| Skill vs. agent vs. hook \| `plugin-entity-taxonomy.md` \|
16	\| Use cases, trigger phrases, test cases \| `skill-building-guidance/use-case-planning.md` \|
17	\| Directory name, file name, dependency prefix \| `skill-building-guidance/naming-conventions.md` \|
18	\| The `description` field (four components, boundaries) \| `skill-building-guidance/skill-description-frontmatter.md`, `skill-building-guidance/skill-description-length.md` \|
19	\| Which frontmatter fields to set \| `skill-building-guidance/skill-frontmatter-fields.md` \|
20	\| Where content lives (body vs. references vs. scripts vs. assets) \| `skill-building-guidance/progressive-disclosure.md`, `skill-building-guidance/skill-reference-files.md` \|
21	\| Step structure and workflow shape \| `skill-building-guidance/workflow-patterns.md`, `skill-building-guidance/writing-effective-instructions.md` \|
22	\| `allowed-tools`, Bash permission granularity \| `skill-building-guidance/allowed-tools-bash-permissions.md`, `skill-building-guidance/allowed-tools-AskUserQuestion.md` \|
23	\| Reading config / runtime data \| `skill-building-guidance/context-injection-commands.md`, `skill-building-guidance/dynamic-project-discovery.md` \|
24	\| Running scripts \| `skill-building-guidance/script-execution-instructions.md` \|
25	\| Dispatching agents from the skill \| `skill-building-guidance/agent-dispatch-namespacing.md`, plus `agent-building-guidelines/multi-agent-economics.md` \|
26	\| Degraded environments (no git, missing tools) \| `skill-building-guidance/graceful-degradation.md`, `skill-building-guidance/optional-git-repositories.md` \|
27	\| Frontmatter safety (angle brackets, YAML types) \| `skill-building-guidance/security-restrictions.md` \|
28	\| Hardening fuzzy steps into deterministic ones \| `skill-building-guidance/hardening-fuzzy-vs-deterministic.md` \|
29	\| Splitting or composing skills \| `skill-building-guidance/skill-decomposition.md`, `skill-building-guidance/skill-composition.md` \|
30	\| Defining success and tests \| `skill-building-guidance/success-criteria-and-testing.md` \|
31	\| New plugin needed (plugin.json, marketplace.json) \| `claude-marketplace-and-plugin-configuration/` and `templates/` \|
32
33	## Operating Principles
34
35	- Interview relentlessly, but explore first. Interview the user relentlessly
36	about every aspect of the skill until you reach a shared understanding. Walk
37	down each branch of the design tree, resolving dependencies between decisions
38	one-by-one. **If a question can be answered by exploring the repository — the
39	target plugin's existing skills, sibling descriptions, `plugin.json`,
40	conventions, the guidance documents above — explore instead of asking.** Only
41	surface questions that genuinely require the user's judgment.
42	- Ask one question at a time. Never batch questions. Settle one decision,
43	let its answer resolve dependent decisions, then ask the next. Later answers
44	routinely make earlier questions moot.
45	- Recommend, then ask. For every question surfaced to the user, provide a
46	recommended answer with rationale grounded in evidence (existing skills,
47	conventions, the guidance, the user's stated goal). The user can accept,
48	amend, or redirect.
49	- Apply guidance as you go, then verify at the end. Consult the governing
50	document when a decision is on the table (Step 4), and run a full
51	guidance-conformance pass over the finished files at the end (Step 6). The
52	interview gets each decision approximately right; the review pass makes the
53	artifact correct.
54
55	# Build a Skill
56
57	## Step 1: Capture the Request and Confirm It Is a Skill
58
59	Read the user's argument and the conversation to extract what the skill should
60	do. If the request is too thin to start (for example, just "build a skill"),
61	ask

People who liked this also viewed

6 picks

Type

skill-builder

122stars12forks

View on GitHub

How to install skill-builder?

testdouble/han/skill-builder

$npx -y skills add testdouble/han --skill skill-builder

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	## Guidance Location
2
3	The authoritative skill-authoring guidance ships in this plugin. Read the
4	specific document a decision needs, when that decision is on the table — never
5	read them all up front, because that defeats progressive disclosure and burns
6	context on guidance the current skill does not touch.
7
8	- Plugin-building guidance root: `${CLAUDE_PLUGIN_ROOT}/skills/guidance/references/`
9	- Skill-specific guidance: `${CLAUDE_PLUGIN_ROOT}/skills/guidance/references/skill-building-guidance/`
10
11	Map from decision to governing document (read just-in-time):
12
13	\| Decision on the table \| Read \|
14	\|---\|---\|
15	\| Skill vs. agent vs. hook \| `plugin-entity-taxonomy.md` \|
16	\| Use cases, trigger phrases, test cases \| `skill-building-guidance/use-case-planning.md` \|
17	\| Directory name, file name, dependency prefix \| `skill-building-guidance/naming-conventions.md` \|
18	\| The `description` field (four components, boundaries) \| `skill-building-guidance/skill-description-frontmatter.md`, `skill-building-guidance/skill-description-length.md` \|
19	\| Which frontmatter fields to set \| `skill-building-guidance/skill-frontmatter-fields.md` \|
20	\| Where content lives (body vs. references vs. scripts vs. assets) \| `skill-building-guidance/progressive-disclosure.md`, `skill-building-guidance/skill-reference-files.md` \|
21	\| Step structure and workflow shape \| `skill-building-guidance/workflow-patterns.md`, `skill-building-guidance/writing-effective-instructions.md` \|
22	\| `allowed-tools`, Bash permission granularity \| `skill-building-guidance/allowed-tools-bash-permissions.md`, `skill-building-guidance/allowed-tools-AskUserQuestion.md` \|
23	\| Reading config / runtime data \| `skill-building-guidance/context-injection-commands.md`, `skill-building-guidance/dynamic-project-discovery.md` \|
24	\| Running scripts \| `skill-building-guidance/script-execution-instructions.md` \|
25	\| Dispatching agents from the skill \| `skill-building-guidance/agent-dispatch-namespacing.md`, plus `agent-building-guidelines/multi-agent-economics.md` \|
26	\| Degraded environments (no git, missing tools) \| `skill-building-guidance/graceful-degradation.md`, `skill-building-guidance/optional-git-repositories.md` \|
27	\| Frontmatter safety (angle brackets, YAML types) \| `skill-building-guidance/security-restrictions.md` \|
28	\| Hardening fuzzy steps into deterministic ones \| `skill-building-guidance/hardening-fuzzy-vs-deterministic.md` \|
29	\| Splitting or composing skills \| `skill-building-guidance/skill-decomposition.md`, `skill-building-guidance/skill-composition.md` \|
30	\| Defining success and tests \| `skill-building-guidance/success-criteria-and-testing.md` \|
31	\| New plugin needed (plugin.json, marketplace.json) \| `claude-marketplace-and-plugin-configuration/` and `templates/` \|
32
33	## Operating Principles
34
35	- Interview relentlessly, but explore first. Interview the user relentlessly
36	about every aspect of the skill until you reach a shared understanding. Walk
37	down each branch of the design tree, resolving dependencies between decisions
38	one-by-one. **If a question can be answered by exploring the repository — the
39	target plugin's existing skills, sibling descriptions, `plugin.json`,
40	conventions, the guidance documents above — explore instead of asking.** Only
41	surface questions that genuinely require the user's judgment.
42	- Ask one question at a time. Never batch questions. Settle one decision,
43	let its answer resolve dependent decisions, then ask the next. Later answers
44	routinely make earlier questions moot.
45	- Recommend, then ask. For every question surfaced to the user, provide a
46	recommended answer with rationale grounded in evidence (existing skills,
47	conventions, the guidance, the user's stated goal). The user can accept,
48	amend, or redirect.
49	- Apply guidance as you go, then verify at the end. Consult the governing
50	document when a decision is on the table (Step 4), and run a full
51	guidance-conformance pass over the finished files at the end (Step 6). The
52	interview gets each decision approximately right; the review pass makes the
53	artifact correct.
54
55	# Build a Skill
56
57	## Step 1: Capture the Request and Confirm It Is a Skill
58
59	Read the user's argument and the conversation to extract what the skill should
60	do. If the request is too thin to start (for example, just "build a skill"),
61	ask

People who liked this also viewed

6 picks

Type