Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions .claude-plugin/marketplace.json
Original file line number Diff line number Diff line change
Expand Up @@ -6,7 +6,7 @@
},
"metadata": {
"description": "Professional AI coding configurations, agents, skills, and context for Claude Code and Cursor",
"version": "9.15.1",
"version": "9.16.0",
"license": "MIT",
"repository": "https://github.com/TechNickAI/ai-coding-config"
},
Expand All @@ -15,7 +15,7 @@
"name": "ai-coding-config",
"source": "./plugins/core",
"description": "Commands, agents, skills, and context for AI-assisted development workflows",
"version": "9.15.1",
"version": "9.16.0",
"tags": ["commands", "agents", "skills", "workflows", "essential"]
}
]
Expand Down
2 changes: 1 addition & 1 deletion plugins/core/.claude-plugin/plugin.json
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
{
"name": "ai-coding-config",
"version": "9.15.1",
"version": "9.16.0",
"description": "Commands, agents, skills, and context for AI-assisted development workflows",
"author": {
"name": "TechNickAI",
Expand Down
78 changes: 78 additions & 0 deletions plugins/core/skills/kungfu/SKILL.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,78 @@
---
name: kungfu
# prettier-ignore
description: "Use when finding new AI agent skills, discovering capabilities, installing skills from GitHub, searching skill marketplaces, or expanding what Claude can do - like Neo downloading martial arts in The Matrix"
version: 1.2.0
category: meta
triggers:
- "kungfu"
- "find skills"
- "install skill"
- "skill marketplace"
- "download skill"
- "new capabilities"
- "agent skills"
- "expand abilities"
- "i know kungfu"
- "matrix"
---

<objective>
Discover, evaluate, and install AI agent skills from the ecosystem. Like Neo downloading kung fu in The Matrix - expand capabilities by finding and integrating new skills.
</objective>

<discovery-strategy>
Search GitHub for skill collections using:
- Repos matching "awesome*skills" or "claude*skills"
- Repos containing SKILL.md files (query: `path:SKILL.md claude`)
- Topics: claude-code, ai-skills, agent-skills

Known curated collections (verify availability before citing):
- VoltAgent/awesome-agent-skills
- composioHQ/awesome-claude-code-skills
- sickn33/antigravity-awesome-skills

If curated lists are unavailable, search GitHub directly. Prioritize repos updated within 6 months with meaningful star counts.
</discovery-strategy>

<skill-format>
Valid skills: SKILL.md with frontmatter (name, "Use when..." description, triggers). Optional scripts/tool for executables (require user approval before install).
</skill-format>

<quality-signals>
**Evaluate:** GitHub stars, recent activity, SKILL.md quality, documentation, relevance to user's need.

**Skip:** No SKILL.md, abandoned (no commits in 1+ year AND unresponsive), <10 stars (unless from trusted source), duplicates already-installed skill.
</quality-signals>

<security>
If a skill includes executable scripts in `scripts/`, ALWAYS:
1. Show the user the script contents
2. Explain what the script does
3. Get explicit approval before installing

Never auto-execute downloaded scripts.
</security>
Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Security review omits executables outside scripts directory

Medium Severity

The <skill-format> section mentions "Optional scripts/tool for executables" suggesting executables can exist in both scripts/ and tool locations. However, the <security> section only instructs to perform security review for "executable scripts in scripts/". This gap could allow a malicious skill to include an executable as a tool file that bypasses the security review process (showing contents, explaining behavior, getting approval).

Additional Locations (1)

Fix in Cursor Fix in Web


<installation>
Install to the project's `.claude/skills/<skill-name>/` directory. For global installation, use `~/.claude/skills/`.

**Before installing:**
- Check for existing skill with same name
- If conflict: ask user to overwrite, rename, or skip

**After downloading:**
- Validate SKILL.md parses without YAML errors
- Confirm required frontmatter (name, description, triggers) exists
- Test with `/skill <name>` or a natural trigger phrase

**If install fails:** Remove any partially downloaded files and report the specific failure.
</installation>

<workflows>
<search>Query GitHub and curated lists, evaluate candidates against quality signals, present top 3-5 options with name, description, star count, last update, and install instructions.</search>

<install>Fetch skill, run security review if scripts present, validate format, install to skills directory, verify it loads.</install>

<audit>List installed skills from `.claude/skills/`, check source repos for updates, identify unused skills for removal.</audit>
Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Audit workflow ignores globally installed skills path

Low Severity

The <installation> section documents two valid paths: project-level (.claude/skills/<skill-name>/) and global (~/.claude/skills/). However, the <audit> workflow only instructs to "List installed skills from .claude/skills/" — it omits the global ~/.claude/skills/ path. Users who install skills globally following the documented instructions won't be able to audit those skills.

Additional Locations (1)

Fix in Cursor Fix in Web

</workflows>