create-project-skills

Name: create-project-skills
Author: tobihagemann

2installs29forks

How to install create-project-skills?

tobihagemann/turbo/create-project-skills

$npx -y skills add tobihagemann/turbo --skill create-project-skills

Installs into the current project.

Summary

Scans an existing codebase and generates project-specific skills that capture inferred conventions such as naming, file organization, framework usage, data access, error handling, and testing style. Writes into the project's chosen skill directory (e.g., .claude/skills/, `.agen

Files · 1

View on GitHub

SKILL.md

1	# Create Project Skills
2
3	Generates one skill per detected convention area in the project's skill directory so future Claude or Codex sessions auto-load them when working in the repo.
4
5	## Task Tracking
6
7	At the start, use `TaskCreate` to create a task for each phase:
8
9	1. Survey codebase
10	2. Extract patterns in parallel
11	3. Evaluate patterns
12	4. Propose skill list
13	5. Run `/create-skill` skill
14
15	## Step 1: Survey Codebase
16
17	If `$ARGUMENTS` specifies paths, scope the scan to those paths; otherwise scan the whole repository.
18
19	Build the extraction context:
20
21	1. Detect primary languages and frameworks from manifest files (`package.json`, `Cargo.toml`, `pyproject.toml`, `go.mod`, `Package.swift`, `pom.xml`, `Gemfile`, and others appropriate to the stack).
22	2. Map the top-level source directory structure and note test directory conventions.
23	3. Read `CLAUDE.md`, `.claude/rules/`, `AGENTS.md`, and any `.cursor/rules` or `.cursorrules`. Note the conventions already documented there. The generated skills must not duplicate them.
24	4. Determine the target skill directory:
25	- Check candidate paths `.claude/skills/` (Claude Code), `.agents/skills/` (Codex), and a top-level `skills/` directory (match case-insensitively so `Skills/` or similar non-standard casing is detected too). Resolve symlinks so co-linked paths are treated as one logical location.
26	- Use `AskUserQuestion` to confirm where generated skills should live. List each distinct resolved location as an option, noting any symlink alias in the option description. If `.claude/skills/` is not among the detected locations, include it as a default option. The auto-added "Other" option lets users specify a custom path such as a project-specific directory.
27	5. In the chosen target directory, list existing skills. For each, record the skill name, the description from SKILL.md frontmatter, and the first `##` section heading from the body. These signals feed rename-conflict detection in Step 3.
28
29	Output a short text summary of detected stack, top-level layout, chosen target directory, and existing skills before moving on.
30
31	## Step 2: Extract Patterns in Parallel
32
33	Read [references/pattern-extractor.md](references/pattern-extractor.md) to see the full taxonomy of pattern categories. Decide which categories apply to the detected stack (e.g., drop "Styling and UI" for a backend service, drop "State management" for a static-analysis tool).
34
35	Use the Agent tool to launch all extraction agents below in a single assistant message so they run concurrently. Run them in the foreground so all their results return in this turn. Each Agent call uses `model: "opus"`. Launch one Agent per applicable category and state the total count explicitly when emitting the calls. Each agent's prompt must:
36
37	- Name its assigned category
38	- Include the stack summary and directory map from Step 1
39	- Include the list of conventions already documented in `CLAUDE.md` and `.claude/rules/` so duplicates are skipped
40	- Instruct the agent to read [references/pattern-extractor.md](references/pattern-extractor.md) as its role brief and return findings in the format defined at the end of that file
41
42	## Step 3: Evaluate Patterns
43
44	Aggregate findings from all agents. For each finding, score three axes:
45
46	- Consistency: what share of eligible sites follow the pattern? Drop findings below 30%. Flag findings between 30–70% as "mixed" for Step 4 review.
47	- Intentionality: does the pattern appear across multiple subsystems and recent commits, or is it isolated? Drop findings confined to a single legacy module unless docs or lint config explicitly mark them as the desired convention.
48	- Modernity: does the pattern align with current best practices for the stack? Flag patterns that contradict current idioms (e.g., pre-hooks class components in a React codebase also using hooks elsewhere) as "legacy" for Step 4 review.
49
50	Group the surviving findings by topic into candidate skills. Each candidate typically covers one category, but related categories may merge if the patterns are tightly coupled. Split a candidate into two skills if its patterns cover clearly distinct sub-topics.
51
52	For each candidate skill, produce:
53
54	- A proposed `name` (kebab-case, narrow to the topic, e.g., `swift-naming`, `react-state`, `api-clients`)
55	- A one-line description with trigger phrases (e.g., "Use when writing or reviewing <topic>...")
56	- 3–8 concrete convention statements with evidence citation

People who liked this also viewed

6 picks

Type

create-project-skills

2installs29forks

View on GitHub

How to install create-project-skills?

tobihagemann/turbo/create-project-skills

$npx -y skills add tobihagemann/turbo --skill create-project-skills

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Create Project Skills
2
3	Generates one skill per detected convention area in the project's skill directory so future Claude or Codex sessions auto-load them when working in the repo.
4
5	## Task Tracking
6
7	At the start, use `TaskCreate` to create a task for each phase:
8
9	1. Survey codebase
10	2. Extract patterns in parallel
11	3. Evaluate patterns
12	4. Propose skill list
13	5. Run `/create-skill` skill
14
15	## Step 1: Survey Codebase
16
17	If `$ARGUMENTS` specifies paths, scope the scan to those paths; otherwise scan the whole repository.
18
19	Build the extraction context:
20
21	1. Detect primary languages and frameworks from manifest files (`package.json`, `Cargo.toml`, `pyproject.toml`, `go.mod`, `Package.swift`, `pom.xml`, `Gemfile`, and others appropriate to the stack).
22	2. Map the top-level source directory structure and note test directory conventions.
23	3. Read `CLAUDE.md`, `.claude/rules/`, `AGENTS.md`, and any `.cursor/rules` or `.cursorrules`. Note the conventions already documented there. The generated skills must not duplicate them.
24	4. Determine the target skill directory:
25	- Check candidate paths `.claude/skills/` (Claude Code), `.agents/skills/` (Codex), and a top-level `skills/` directory (match case-insensitively so `Skills/` or similar non-standard casing is detected too). Resolve symlinks so co-linked paths are treated as one logical location.
26	- Use `AskUserQuestion` to confirm where generated skills should live. List each distinct resolved location as an option, noting any symlink alias in the option description. If `.claude/skills/` is not among the detected locations, include it as a default option. The auto-added "Other" option lets users specify a custom path such as a project-specific directory.
27	5. In the chosen target directory, list existing skills. For each, record the skill name, the description from SKILL.md frontmatter, and the first `##` section heading from the body. These signals feed rename-conflict detection in Step 3.
28
29	Output a short text summary of detected stack, top-level layout, chosen target directory, and existing skills before moving on.
30
31	## Step 2: Extract Patterns in Parallel
32
33	Read [references/pattern-extractor.md](references/pattern-extractor.md) to see the full taxonomy of pattern categories. Decide which categories apply to the detected stack (e.g., drop "Styling and UI" for a backend service, drop "State management" for a static-analysis tool).
34
35	Use the Agent tool to launch all extraction agents below in a single assistant message so they run concurrently. Run them in the foreground so all their results return in this turn. Each Agent call uses `model: "opus"`. Launch one Agent per applicable category and state the total count explicitly when emitting the calls. Each agent's prompt must:
36
37	- Name its assigned category
38	- Include the stack summary and directory map from Step 1
39	- Include the list of conventions already documented in `CLAUDE.md` and `.claude/rules/` so duplicates are skipped
40	- Instruct the agent to read [references/pattern-extractor.md](references/pattern-extractor.md) as its role brief and return findings in the format defined at the end of that file
41
42	## Step 3: Evaluate Patterns
43
44	Aggregate findings from all agents. For each finding, score three axes:
45
46	- Consistency: what share of eligible sites follow the pattern? Drop findings below 30%. Flag findings between 30–70% as "mixed" for Step 4 review.
47	- Intentionality: does the pattern appear across multiple subsystems and recent commits, or is it isolated? Drop findings confined to a single legacy module unless docs or lint config explicitly mark them as the desired convention.
48	- Modernity: does the pattern align with current best practices for the stack? Flag patterns that contradict current idioms (e.g., pre-hooks class components in a React codebase also using hooks elsewhere) as "legacy" for Step 4 review.
49
50	Group the surviving findings by topic into candidate skills. Each candidate typically covers one category, but related categories may merge if the patterns are tightly coupled. Split a candidate into two skills if its patterns cover clearly distinct sub-topics.
51
52	For each candidate skill, produce:
53
54	- A proposed `name` (kebab-case, narrow to the topic, e.g., `swift-naming`, `react-state`, `api-clients`)
55	- A one-line description with trigger phrases (e.g., "Use when writing or reviewing <topic>...")
56	- 3–8 concrete convention statements with evidence citation

People who liked this also viewed

6 picks

Type