publish-skill

Name: publish-skill
Author: Aperivue

208stars53forks

How to install publish-skill?

aperivue/medsci-skills/publish-skill

$npx -y skills add Aperivue/medsci-skills --skill publish-skill

Installs into the current project.

Summary

Convert a personal agent skill into a distributable, open-source-ready skill. Runs PII audit, generalization, license compatibility check, cross-platform adapter review, and packaging workflow.

Files · 1

View on GitHub

SKILL.md

1	# Skill: publish-skill
2
3	Convert a personal agent skill into a clean, distributable, open-source-ready skill package. This skill walks through a 7-phase pipeline that audits for personally identifiable information, generalizes language and role assumptions, verifies license compatibility, checks cross-platform adapter needs, and prepares the final package for commit.
4
5	## Communication Rules
6
7	- Communicate with the user in their preferred language
8	- Use English for technical terms (PII, MIT, CC BY, GPL, YAML frontmatter)
9	- Present audit findings in structured tables
10
11	---
12
13	## Phase 0: Init and Identify Source
14
15	### Required Inputs
16
17	Collect from the user:
18
19	1. Source skill path: directory containing the personal skill (e.g., `~/.claude/skills/my-skill/` or `~/.agents/skills/my-skill/`)
20	2. Target package path: directory of the distributable package (e.g., `~/workspace/<your-package>/`)
21	3. Target license: license of the package (default: MIT)
22
23	### Actions
24
25	1. Read `SKILL.md` from the source skill directory
26	2. Inventory all files recursively (`ls -R`)
27	3. Classify skill type:
28	- Standalone: self-contained skill with no agent delegation
29	- Orchestrator: delegates to sub-agents (NOT suitable for distribution without refactoring)
30	- Wrapper: thin wrapper around another tool/API
31	4. Present inventory table to the user:
32
33	```
34	\| File \| Lines \| Type \| Notes \|
35	\|------\|-------\|------\|-------\|
36	```
37
38	Gate: User confirms source skill and target package before proceeding.
39
40	---
41
42	## Phase 0.5: Skill-Worthiness Gate
43
44	Before spending effort on PII scrubbing and generalization, confirm the workflow is
45	worth distributing as a skill at all. A skill earns its place by encoding a reusable
46	decision heuristic, a hard-won constraint, or a verification step — not a snippet anyone
47	could reconstruct in five minutes. Apply all three gates; **any "no" (or "yes" on the
48	inverse) stops publication** in favor of documentation or a memory note instead.
49
50	\| Gate \| Question \| Pass condition \|
51	\|------\|----------\|----------------\|
52	\| Uniqueness \| Could a competent user get the same result by searching the web for ~5 minutes, or by asking a general assistant with no skill installed? \| No \|
53	\| Specificity \| Does it encode a workflow, decision heuristic, constraint, or convention specific to this domain or a recurring task — rather than a generic code snippet or a standard-library example? \| Yes \|
54	\| Effort \| Did discovering it take real debugging, study design, operational effort, or a reviewer-anticipation lesson (a pitfall, a verification step, a domain convention)? \| Yes \|
55
56	Favor skills that encode reviewer-anticipation, reporting-guideline constraints,
57	verification gates, and decision trees over thin wrappers or one-off snippets. This is
58	the publish-time analogue of the "reusable pattern vs one-off hack" distinction: a
59	workflow that fails the gate is better captured as a doc or a memory note than shipped as
60	a skill that dilutes the catalog.
61
62	Gate: If any of the three fails, recommend documentation/memory instead and stop. If
63	the value is real but the skill delegates to private agents, route through Phase 1's
64	orchestrator finding (refactor to standalone first). Only a clear three-way pass proceeds.
65
66	---
67
68	## Phase 1: Originality Check
69
70	Verify the skill is original work suitable for open-source distribution.
71
72	### Checks
73
74	1. External source: Is this skill adapted from another package or author? Check for attribution headers, license blocks, or "based on" comments.
75	2. Third-party content: Do any files in `references/` come from external sources (published guidelines, textbooks, standards bodies)?
76	3. Competitive sensitivity: Does the skill reveal proprietary business logic or competitive advantage that should remain private?
77
78	### Decision Matrix
79
80	\| Finding \| Action \|
81	\|---------\|--------\|
82	\| Fully original \| Proceed to Phase 2 \|
83	\| Adapted with compatible license \| Add attribution header, proceed \|
84	\| Contains non-compatible third-party content \| Flag for removal or URL manifest conversion \|
85	\| Orchestrator with private agent references \| STOP -- requires refactoring to standalone first \|
86	\| Competitive/proprietary logic \| STOP -- not suitable for open-source \|
87
88	---
89
90	## Phase 2: PII De-identification Audit
91
92	Zero tolerance: the skill must have exactly 0 PII matches before proceeding.
93
94	### Pre-scan Setup
95
96	Before running, ask the user for everything that should also count as a PII hit but is unique to them:
97	- Their name(s) in all languages and romanizations (a placeholder shape: `<First Last>\|<native-script

People who liked this also viewed

6 picks

Type

hyperframes-cliUse the HyperFrames CLI development loop: init, add, catalog, capture, lint, check, snapshot, compare, grade-compare, preview, play, present, beats, keyframes, single or batch render, publish, cloud, SkillsJul 2026262k37k
request-refactor-planSkills for Real Engineers. Straight from my .agents directory.SkillsJul 2026165k181k
analyze-projectRigor Analyze / Rigor Audit read-only skill for deep learning research repositories. Use when the user wants to read and understand a repository, inspect model structure and training or inference entrSkillsMay 2026161k507
explore-codeRigor Improve implementation leaf skill for auditable candidate implementation in deep learning research repositories. Use when the researcher explicitly authorizes exploratory work on an isolated braSkillsMay 2026161k507
ubiquitous-languageSkills for Real Engineers. Straight from my .agents directory.SkillsJul 2026160k181k
scaffold-exercisesSkills for Real Engineers. Straight from my .agents directory.SkillsJul 2026149k181k

publish-skill

208stars53forks

View on GitHub

How to install publish-skill?

aperivue/medsci-skills/publish-skill

$npx -y skills add Aperivue/medsci-skills --skill publish-skill

Installs into the current project.

Summary

Convert a personal agent skill into a distributable, open-source-ready skill. Runs PII audit, generalization, license compatibility check, cross-platform adapter review, and packaging workflow.

Files · 1

View on GitHub

SKILL.md

1	# Skill: publish-skill
2
3	Convert a personal agent skill into a clean, distributable, open-source-ready skill package. This skill walks through a 7-phase pipeline that audits for personally identifiable information, generalizes language and role assumptions, verifies license compatibility, checks cross-platform adapter needs, and prepares the final package for commit.
4
5	## Communication Rules
6
7	- Communicate with the user in their preferred language
8	- Use English for technical terms (PII, MIT, CC BY, GPL, YAML frontmatter)
9	- Present audit findings in structured tables
10
11	---
12
13	## Phase 0: Init and Identify Source
14
15	### Required Inputs
16
17	Collect from the user:
18
19	1. Source skill path: directory containing the personal skill (e.g., `~/.claude/skills/my-skill/` or `~/.agents/skills/my-skill/`)
20	2. Target package path: directory of the distributable package (e.g., `~/workspace/<your-package>/`)
21	3. Target license: license of the package (default: MIT)
22
23	### Actions
24
25	1. Read `SKILL.md` from the source skill directory
26	2. Inventory all files recursively (`ls -R`)
27	3. Classify skill type:
28	- Standalone: self-contained skill with no agent delegation
29	- Orchestrator: delegates to sub-agents (NOT suitable for distribution without refactoring)
30	- Wrapper: thin wrapper around another tool/API
31	4. Present inventory table to the user:
32
33	```
34	\| File \| Lines \| Type \| Notes \|
35	\|------\|-------\|------\|-------\|
36	```
37
38	Gate: User confirms source skill and target package before proceeding.
39
40	---
41
42	## Phase 0.5: Skill-Worthiness Gate
43
44	Before spending effort on PII scrubbing and generalization, confirm the workflow is
45	worth distributing as a skill at all. A skill earns its place by encoding a reusable
46	decision heuristic, a hard-won constraint, or a verification step — not a snippet anyone
47	could reconstruct in five minutes. Apply all three gates; **any "no" (or "yes" on the
48	inverse) stops publication** in favor of documentation or a memory note instead.
49
50	\| Gate \| Question \| Pass condition \|
51	\|------\|----------\|----------------\|
52	\| Uniqueness \| Could a competent user get the same result by searching the web for ~5 minutes, or by asking a general assistant with no skill installed? \| No \|
53	\| Specificity \| Does it encode a workflow, decision heuristic, constraint, or convention specific to this domain or a recurring task — rather than a generic code snippet or a standard-library example? \| Yes \|
54	\| Effort \| Did discovering it take real debugging, study design, operational effort, or a reviewer-anticipation lesson (a pitfall, a verification step, a domain convention)? \| Yes \|
55
56	Favor skills that encode reviewer-anticipation, reporting-guideline constraints,
57	verification gates, and decision trees over thin wrappers or one-off snippets. This is
58	the publish-time analogue of the "reusable pattern vs one-off hack" distinction: a
59	workflow that fails the gate is better captured as a doc or a memory note than shipped as
60	a skill that dilutes the catalog.
61
62	Gate: If any of the three fails, recommend documentation/memory instead and stop. If
63	the value is real but the skill delegates to private agents, route through Phase 1's
64	orchestrator finding (refactor to standalone first). Only a clear three-way pass proceeds.
65
66	---
67
68	## Phase 1: Originality Check
69
70	Verify the skill is original work suitable for open-source distribution.
71
72	### Checks
73
74	1. External source: Is this skill adapted from another package or author? Check for attribution headers, license blocks, or "based on" comments.
75	2. Third-party content: Do any files in `references/` come from external sources (published guidelines, textbooks, standards bodies)?
76	3. Competitive sensitivity: Does the skill reveal proprietary business logic or competitive advantage that should remain private?
77
78	### Decision Matrix
79
80	\| Finding \| Action \|
81	\|---------\|--------\|
82	\| Fully original \| Proceed to Phase 2 \|
83	\| Adapted with compatible license \| Add attribution header, proceed \|
84	\| Contains non-compatible third-party content \| Flag for removal or URL manifest conversion \|
85	\| Orchestrator with private agent references \| STOP -- requires refactoring to standalone first \|
86	\| Competitive/proprietary logic \| STOP -- not suitable for open-source \|
87
88	---
89
90	## Phase 2: PII De-identification Audit
91
92	Zero tolerance: the skill must have exactly 0 PII matches before proceeding.
93
94	### Pre-scan Setup
95
96	Before running, ask the user for everything that should also count as a PII hit but is unique to them:
97	- Their name(s) in all languages and romanizations (a placeholder shape: `<First Last>\|<native-script

People who liked this also viewed

6 picks

Type