skills-hub
Central skill/knowledge registry that syncs with kjuhwa/skills.git.
Skill by kjuhwa
Chalk Labs
Need help deploying skills-hub?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/kjuhwa-skills-hub
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/skills-hub
What's inside
- Skill name
skills-hub- Description
- Central skill/knowledge registry that syncs with kjuhwa/skills.git. Discover, install, extract, publish, and maintain reusable skills across projects. Auto-loads when any /init_skills, /skills_*, /skill_extract, /skill_publish, /skill_finalize command is invoked, or when the user mentions installing skills, publishing skills, syncing the skill repo, or extracting skills from the current project.
- License
- MIT
- Instructions
- 118 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# Skills Hub
Central orchestration layer for the `kjuhwa/skills` knowledge repository.
## Remote Repository
- URL: `https://github.com/kjuhwa/skills.git`
- Clone cache: `~/.claude/skills-hub/remote/` (shallow, auto-refreshed)
- Registry: `~/.claude/skills-hub/registry.json` — tracks installed skills, source path, version, install date
## Remote Layout (category-separated)
```
skills/
<category>/ # apm, backend, frontend, devops, testing, db, security, ai, ...
<skill-name>/
SKILL.md # frontmatter: name, description, tags, triggers, category, version
content.md # prompt / knowledge body
examples/ # optional
index.json # flat catalog built from SKILL.md frontmatter
CATEGORIES.md # human-readable category list
README.md
```
## Local Install Destinations
- **Project-scoped**: `.claude/skills/<skill-name>/` (default when inside a project with `.claude/`)
- **User-scoped**: `~/.claude/skills/<skill-name>/` (fallback, or `--global` flag)
Registry entry format:
```json
{
"<skill-name>": {
"category": "apm",
"source_commit": "a1b2c3",
"installed_at": "2026-04-14T10:00:00Z",
"synced_at": "2026-04-15T09:30:00Z",
"scope": "project|global",
"version": "1.0.0",
"pinned": false
}
}
```
## Versioning Contract (git tags)
- Every published skill version gets an annotated tag: `skills/<name>/v<semver>` (e.g. `skills/foo/v1.2.0`).
- `/skills_publish` creates and pushes the tag as part of the publish flow; bumps the version in SKILL.md frontmatter if the existing tag is unchanged.
- `/init_skills foo@1.2.0` (or `--version=1.2.0`) installs the files at that tag and sets `pinned: true` in the registry.
- `/skills_sync --skill=foo --version=1.1.0` performs a rollback to that tag (also pins).
- `/skills_sync --skill=foo --unpin` clears the pin so the skill tracks latest again.
- Bulk `/skills_sync` skips pinned skills unless `--force`.
## Command Map
| Slash Command | Purpose |
|---|---|
| `/init_skills <keyword>` | Search remote by keyword/category → interactive install |
| `/skills_search <keyword>` | Preview-only remote search |
| `/skills_list` | Show installed skills (local + scope + source) |
| `/skills_sync` | Refresh remote cache + update installed skills |
| `/skills_extract` | **Full project scan** → generate skill drafts in `.skills-draft/` |
| `/skills_extract_session` | **Current session only** → drafts from recent changes |
| `/skills_publish` | Review drafts → push to remote as branch + optional PR |
| `/skills_merge <selectors...>` | Combine 2+ remote skills/knowledge into one new draft (read-only on remote) |
| `/skills_split <selector>` | Decompose one remote skill/knowledge into N smaller drafts (read-only on remote) |
| `/skills_refactor` | Scan remote for merge + split candidates and produce both kinds of drafts in one pass |
| `/skills_finalize` | End-of-project: extract → review → publish → cleanup drafts |
| `/skills_cleanup` | Remote maintenance: dedupe, re-index, stale removal (dry-run default) |
| `/skills_remove <name>` | Uninstall local skill |
| `/skills_bootstrap_update [--version=x.y.z]` | Update slash-command files themselves to latest or a tagged version |
| `/skills_bootstrap_publish [--bump=…]` | Publish local command edits to remote with a `bootstrap/v<ver>` tag |
## Safety Rules
1. **Never push directly to `main`** — always a branch `skills/<category>/<name>-<date>`.
2. `publish` and `cleanup` run dry-run first, require explicit user approval before git write operations.
3. `extract` writes to `.skills-draft/` (gitignored by default), never directly to remote.
4. Skill names must match `[a-z0-9-]+`; categories from `CATEGORIES.md` (new ones require explicit approval).
5. Before installing, check registry for name collision; prompt on conflict.
## Extract Heuristics
Only keep patterns that are **generalizable**. Drop project-specific code.
Signals that promote a pattern to a skill draft:
- Appears across multiple files with small variations
- Solves a recurring problem documented in commit messages
- User explicitly corrects behavior (see feedback memory)
- New tool/library integration with non-trivial setup
- Debugging workflow that succeeded after iteration
Signals that demote (skip):
- Contains business names, credentials, internal URLs
- Tied to one-off migration or hotfix
- Reproduces framework defaults
## Metadata Contract (SKILL.md frontmatter)
```yaml
---
name: <kebab-case-name>
description: <one sentence, specific, trigger-friendly>
category: <from CATEGORIES.md>
tags: [list, of, keywords]
triggers: [keyword patterns for auto-activation]
source_project: <originating project name or "manual">
version: 1.0.0
---
```
## Integration Notes
- Works alongside OMC skills — installed skills appear under `~/.claude/skills/` or `.claude/skills/` and are auto-discovered by Claude Code's skill loader.
- Respects `DISABLE_OMC` and `OMC_SKIP_HOOKS`.
- All git operations use `git -C ~/.claude/skills-hub/remote` — never touch the user's project git state.From kjuhwa/skills-hub, MIT.
More from kjuhwa/skills-hub
- 0.406score
credential-cache-file-for-mcp-subprocess
kjuhwa
Write decrypted per-source credentials to a session-scoped .credential-cache.json on disk so MCP stdio subprocesses can read them without the parent passing…
SkillDFreshness: DClaude Code +9 - 0.406score
langsmith-traceable-node-decorator-pattern
kjuhwa
Wrap each LangGraph node function with @traceable(name=...) so every step appears as a discrete child run in LangSmith, plus emit semantic LLM run_names via…
SkillDFreshness: DClaude Code +9 - 0.406score
llm-empty-plan-controller-fallback
kjuhwa
When a planning LLM returns an empty action list, force a deterministic verification action from a priority-ordered fallback list so the agent never spins on…
SkillDFreshness: DClaude Code +9 - 0.406score
alert-time-window-with-buffer
kjuhwa
Compute a query time window that covers the alert plus a 30-min buffer by parsing startsAt from various schemas (Alertmanager nested, top-level…
SkillDFreshness: DClaude Code +9 - 0.406score
anthropic-compatible-endpoint-swap
kjuhwa
Point the Claude Agent SDK at any Anthropic-compatible endpoint (OpenRouter, Vercel AI Gateway, Ollama, custom) by setting ANTHROPIC_BASE_URL +…
SkillDFreshness: DClaude Code +9 - 0.406score
bash-ast-permission-validator
kjuhwa
Validate agent bash commands by parsing them into an AST (bash-parser) and evaluating each subcommand of a pipeline/&&/|| chain against an allowlist, instead…
SkillDFreshness: DClaude Code +9 - 0.406score
claude-agent-sdk-multi-provider-bridge
kjuhwa
Drive non-Anthropic providers (OpenRouter, Ollama, Vercel AI Gateway, any Anthropic-compatible endpoint) through the Claude Agent SDK by swapping base URL +…
SkillDFreshness: DClaude Code +9 - 0.406score
evidence-driven-claim-validator
kjuhwa
Reject LLM-generated root-cause claims that aren't supported by the actual evidence dict — claim mentions "memory" but no host_metrics?
SkillDFreshness: DClaude Code +9 - 0.406score
fetch-interceptor-via-require-preload
kjuhwa
Patch globalThis.fetch in SDK subprocesses BEFORE the SDK captures it, by shipping a CJS interceptor bundle and loading it with node --require / bun --preload…
SkillDFreshness: DClaude Code +9 - 0.406score
healthy-short-circuit-skip-llm
kjuhwa
Bypass an expensive LLM diagnosis when an alert is provably healthy (resolved/info severity, no error annotations, evidence-investigated-and-empty), guarded…
SkillDFreshness: DClaude Code +9 - 0.406score
json-schema-from-python-signature
kjuhwa
Generate a JSON Schema for an LLM tool's input from a Python function signature using inspect + get_type_hints, mapping str→string, int→integer, list→array…
SkillDFreshness: DClaude Code +9 - 0.406score
langgraph-auth-inject-from-runnable-config
kjuhwa
Hydrate auth/user/org/thread/run identifiers into LangGraph state from the RunnableConfig at the very first node, so every downstream node sees a uniform…
SkillDFreshness: DClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 17
- Last commit (this folder)
- Apr 15, 2026 (6 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/kjuhwa/skills-hub
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from kjuhwa
- 0.406score
credential-cache-file-for-mcp-subprocess
kjuhwa
Write decrypted per-source credentials to a session-scoped .credential-cache.json on disk so MCP stdio subprocesses can read them without the parent passing…
SkillDFreshness: DClaude Code +9 - 0.406score
langsmith-traceable-node-decorator-pattern
kjuhwa
Wrap each LangGraph node function with @traceable(name=...) so every step appears as a discrete child run in LangSmith, plus emit semantic LLM run_names via…
SkillDFreshness: DClaude Code +9 - 0.406score
llm-empty-plan-controller-fallback
kjuhwa
When a planning LLM returns an empty action list, force a deterministic verification action from a priority-ordered fallback list so the agent never spins on…
SkillDFreshness: DClaude Code +9 - 0.406score
alert-time-window-with-buffer
kjuhwa
Compute a query time window that covers the alert plus a 30-min buffer by parsing startsAt from various schemas (Alertmanager nested, top-level…
SkillDFreshness: DClaude Code +9 - 0.406score
anthropic-compatible-endpoint-swap
kjuhwa
Point the Claude Agent SDK at any Anthropic-compatible endpoint (OpenRouter, Vercel AI Gateway, Ollama, custom) by setting ANTHROPIC_BASE_URL +…
SkillDFreshness: DClaude Code +9 - 0.406score
bash-ast-permission-validator
kjuhwa
Validate agent bash commands by parsing them into an AST (bash-parser) and evaluating each subcommand of a pipeline/&&/|| chain against an allowlist, instead…
SkillDFreshness: DClaude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get skills-hub running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
skills-hub. Agent Mart. https://agentmart.dev/skills/kjuhwa-skills-hub (last updated 2026-10-24)