Skip to content

aiskillstore/marketplace / skills/92bilal26/skill-creator

skill-creator

Verified

Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Claude's capabilities with specialized knowledge, workflows, or tool integrations.

FileKindSize
LICENSE.txtmarkdown11.1 KB
SKILL.mdskill md11.3 KB
scripts/init_skill.pyscript10.6 KB
scripts/package_skill.pyscript3.2 KB
scripts/quick_validate.pyscript2.1 KB
skill-report.jsonconfig44.3 KB

skill-report.json

View on GitHub
{
  "schema_version": "2.0",
  "meta": {
    "generated_at": "2026-07-05T16:11:28.099Z",
    "slug": "92bilal26-skill-creator",
    "source_url": "https://github.com/92Bilal26/TaskPilotAI/tree/main/.claude/skills/skill-creator",
    "source_ref": "main",
    "model": "codex",
    "analysis_version": "3.0.0",
    "source_type": "community",
    "content_hash": "8c0ce23bb87be91f5b0853c2e9815617c3858b63acdddfe5da47d2c2d3a60816",
    "tree_hash": "4cc6c16132217766c029dc3003d03831083a0e9c03a4dd4b7195919759ba4510",
    "provenance": {
      "model_effective": "codex",
      "fallback_chain": [
        {
          "agent": "codex",
          "outcome": "success"
        }
      ],
      "analysis_version": "3.0.0"
    }
  },
  "skill": {
    "name": "skill-creator",
    "description": "Guides AI assistants through creating and updating reusable skills with clear metadata, structured instructions, optional resources, validation, and packaging.",
    "summary": "Create and package reusable AI skills with structured guidance, helper scripts, and validation.",
    "icon": "📦",
    "version": "1.0.1",
    "author": "92Bilal26",
    "license": "Complete terms in LICENSE.txt",
    "tags": [
      "ai-skills",
      "skill-authoring",
      "documentation",
      "validation",
      "packaging"
    ],
    "supported_tools": [
      "claude",
      "codex",
      "claude-code"
    ],
    "risk_factors": [
      "external_commands",
      "filesystem"
    ],
    "category": "coding"
  },
  "security_audit": {
    "risk_level": "medium",
    "is_blocked": false,
    "safe_to_publish": true,
    "analysis_status": "ok",
    "summary": "Most static detections are false positives caused by Markdown backticks and documentation examples, not Ruby or shell backtick execution. The confirmed risks are bundled helper commands that write local files, unfiltered recursive packaging, and missing validation for path-like skill names. No prompt injection attempt was found in the reviewed skill files.",
    "remediation": [
      {
        "issue": "The initializer accepts path-like skill names before creating files.",
        "suggestion": "Validate skill_name against the documented hyphen-case pattern and verify the resolved skill directory remains under the requested base path.",
        "severity": "medium"
      },
      {
        "issue": "Recursive packaging may include unintended files or symlink targets.",
        "suggestion": "Exclude hidden files, credentials, caches, VCS directories, generated archives, and symlinks before writing the zip archive.",
        "severity": "medium"
      },
      {
        "issue": "SKILL.md instructs agents to run bundled scripts that write local files.",
        "suggestion": "Document expected inputs, output paths, and file effects directly before each command example.",
        "severity": "medium"
      },
      {
        "issue": "The packager can write the output archive into a broad or unintended directory.",
        "suggestion": "Require an explicit output directory or warn before writing archives outside a project-controlled dist folder.",
        "severity": "low"
      }
    ],
    "risk_factor_evidence": [
      {
        "factor": "external_commands",
        "evidence": [
          {
            "file": "scripts/init_skill.py",
            "line_start": 73,
            "line_end": 73
          },
          {
            "file": "scripts/init_skill.py",
            "line_start": 74,
            "line_end": 74
          },
          {
            "file": "scripts/init_skill.py",
            "line_start": 84,
            "line_end": 84
          },
          {
            "file": "SKILL.md",
            "line_start": 29,
            "line_end": 40
          },
          {
            "file": "SKILL.md",
            "line_start": 40,
            "line_end": 44
          },
          {
            "file": "SKILL.md",
            "line_start": 44,
            "line_end": 44
          },
          {
            "file": "SKILL.md",
            "line_start": 48,
            "line_end": 53
          },
          {
            "file": "SKILL.md",
            "line_start": 53,
            "line_end": 57
          },
          {
            "file": "SKILL.md",
            "line_start": 57,
            "line_end": 62
          },
          {
            "file": "SKILL.md",
            "line_start": 62,
            "line_end": 62
          },
          {
            "file": "SKILL.md",
            "line_start": 68,
            "line_end": 73
          },
          {
            "file": "SKILL.md",
            "line_start": 73,
            "line_end": 73
          },
          {
            "file": "SKILL.md",
            "line_start": 115,
            "line_end": 118
          },
          {
            "file": "SKILL.md",
            "line_start": 118,
            "line_end": 120
          },
          {
            "file": "SKILL.md",
            "line_start": 120,
            "line_end": 123
          },
          {
            "file": "SKILL.md",
            "line_start": 123,
            "line_end": 125
          },
          {
            "file": "SKILL.md",
            "line_start": 125,
            "line_end": 128
          },
          {
            "file": "SKILL.md",
            "line_start": 128,
            "line_end": 138
          },
          {
            "file": "SKILL.md",
            "line_start": 138,
            "line_end": 142
          },
          {
            "file": "SKILL.md",
            "line_start": 142,
            "line_end": 144
          },
          {
            "file": "SKILL.md",
            "line_start": 144,
            "line_end": 150
          },
          {
            "file": "SKILL.md",
            "line_start": 150,
            "line_end": 150
          },
          {
            "file": "SKILL.md",
            "line_start": 161,
            "line_end": 161
          },
          {
            "file": "SKILL.md",
            "line_start": 163,
            "line_end": 163
          },
          {
            "file": "SKILL.md",
            "line_start": 179,
            "line_end": 181
          },
          {
            "file": "SKILL.md",
            "line_start": 181,
            "line_end": 185
          },
          {
            "file": "SKILL.md",
            "line_start": 185,
            "line_end": 187
          },
          {
            "file": "SKILL.md",
            "line_start": 187,
            "line_end": 197
          }
        ]
      },
      {
        "factor": "filesystem",
        "evidence": [
          {
            "file": "scripts/package_skill.py",
            "line_start": 14,
            "line_end": 14
          },
          {
            "file": "scripts/package_skill.py",
            "line_start": 68,
            "line_end": 68
          }
        ]
      }
    ],
    "static_findings": [
      {
        "id": "external_commands:scripts/init_skill.py:73:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "scripts/init_skill.py",
        "line_start": 73,
        "line_end": 73,
        "snippet": "- PDF skill: `fill_fillable_fields.py`, `extract_form_field_info.py` - utilities for PDF manipulatio"
      },
      {
        "id": "external_commands:scripts/init_skill.py:74:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "scripts/init_skill.py",
        "line_start": 74,
        "line_end": 74,
        "snippet": "- DOCX skill: `document.py`, `utilities.py` - Python modules for document processing"
      },
      {
        "id": "external_commands:scripts/init_skill.py:84:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "scripts/init_skill.py",
        "line_start": 84,
        "line_end": 84,
        "snippet": "- Product management: `communication.md`, `context_building.md` - detailed workflow guides"
      },
      {
        "id": "filesystem:scripts/package_skill.py:14:python-archive-libraries",
        "category": "filesystem",
        "severity": "medium",
        "pattern": "Python archive libraries",
        "file": "scripts/package_skill.py",
        "line_start": 14,
        "line_end": 14,
        "snippet": "import zipfile"
      },
      {
        "id": "filesystem:scripts/package_skill.py:68:python-archive-libraries",
        "category": "filesystem",
        "severity": "medium",
        "pattern": "Python archive libraries",
        "file": "scripts/package_skill.py",
        "line_start": 68,
        "line_end": 68,
        "snippet": "with zipfile.ZipFile(zip_filename, 'w', zipfile.ZIP_DEFLATED) as zipf:"
      },
      {
        "id": "blocker:scripts/quick_validate.py:28:system-reconnaissance",
        "category": "blocker",
        "severity": "low",
        "pattern": "System reconnaissance",
        "file": "scripts/quick_validate.py",
        "line_start": 28,
        "line_end": 28,
        "snippet": "return False, \"Invalid frontmatter format\""
      },
      {
        "id": "blocker:scripts/quick_validate.py:65:system-reconnaissance",
        "category": "blocker",
        "severity": "low",
        "pattern": "System reconnaissance",
        "file": "scripts/quick_validate.py",
        "line_start": 65,
        "line_end": 65,
        "snippet": "sys.exit(0 if valid else 1)"
      },
      {
        "id": "external_commands:SKILL.md:29:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 29,
        "line_end": 40,
        "snippet": "```"
      },
      {
        "id": "external_commands:SKILL.md:40:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 40,
        "line_end": 44,
        "snippet": "```"
      },
      {
        "id": "external_commands:SKILL.md:44:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 44,
        "line_end": 44,
        "snippet": "**Metadata Quality:** The `name` and `description` in YAML frontmatter determine when Claude will us"
      },
      {
        "id": "external_commands:SKILL.md:48:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 48,
        "line_end": 53,
        "snippet": "##### Scripts (`scripts/`)"
      },
      {
        "id": "external_commands:SKILL.md:53:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 53,
        "line_end": 57,
        "snippet": "- **Example**: `scripts/rotate_pdf.py` for PDF rotation tasks"
      },
      {
        "id": "external_commands:SKILL.md:57:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 57,
        "line_end": 62,
        "snippet": "##### References (`references/`)"
      },
      {
        "id": "external_commands:SKILL.md:62:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 62,
        "line_end": 62,
        "snippet": "- **Examples**: `references/finance.md` for financial schemas, `references/mnda.md` for company NDA "
      },
      {
        "id": "external_commands:SKILL.md:68:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 68,
        "line_end": 73,
        "snippet": "##### Assets (`assets/`)"
      },
      {
        "id": "external_commands:SKILL.md:73:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 73,
        "line_end": 73,
        "snippet": "- **Examples**: `assets/logo.png` for brand assets, `assets/slides.pptx` for PowerPoint templates, `"
      },
      {
        "id": "external_commands:SKILL.md:115:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 115,
        "line_end": 118,
        "snippet": "Example: When building a `pdf-editor` skill to handle queries like \"Help me rotate this PDF,\" the an"
      },
      {
        "id": "external_commands:SKILL.md:118:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 118,
        "line_end": 120,
        "snippet": "2. A `scripts/rotate_pdf.py` script would be helpful to store in the skill"
      },
      {
        "id": "external_commands:SKILL.md:120:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 120,
        "line_end": 123,
        "snippet": "Example: When designing a `frontend-webapp-builder` skill for queries like \"Build me a todo app\" or "
      },
      {
        "id": "external_commands:SKILL.md:123:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 123,
        "line_end": 125,
        "snippet": "2. An `assets/hello-world/` template containing the boilerplate HTML/React project files would be he"
      },
      {
        "id": "external_commands:SKILL.md:125:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 125,
        "line_end": 128,
        "snippet": "Example: When building a `big-query` skill to handle queries like \"How many users have logged in tod"
      },
      {
        "id": "external_commands:SKILL.md:128:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 128,
        "line_end": 138,
        "snippet": "2. A `references/schema.md` file documenting the table schemas would be helpful to store in the skil"
      },
      {
        "id": "external_commands:SKILL.md:138:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 138,
        "line_end": 142,
        "snippet": "When creating a new skill from scratch, always run the `init_skill.py` script. The script convenient"
      },
      {
        "id": "external_commands:SKILL.md:142:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 142,
        "line_end": 144,
        "snippet": "```bash"
      },
      {
        "id": "external_commands:SKILL.md:144:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 144,
        "line_end": 150,
        "snippet": "```"
      },
      {
        "id": "external_commands:SKILL.md:150:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 150,
        "line_end": 150,
        "snippet": "- Creates example resource directories: `scripts/`, `references/`, and `assets/`"
      },
      {
        "id": "external_commands:SKILL.md:161:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 161,
        "line_end": 161,
        "snippet": "To begin implementation, start with the reusable resources identified above: `scripts/`, `references"
      },
      {
        "id": "external_commands:SKILL.md:163:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 163,
        "line_end": 163,
        "snippet": "Also, delete any example files and directories not needed for the skill. The initialization script c"
      },
      {
        "id": "external_commands:SKILL.md:179:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 179,
        "line_end": 181,
        "snippet": "```bash"
      },
      {
        "id": "external_commands:SKILL.md:181:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 181,
        "line_end": 185,
        "snippet": "```"
      },
      {
        "id": "external_commands:SKILL.md:185:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 185,
        "line_end": 187,
        "snippet": "```bash"
      },
      {
        "id": "external_commands:SKILL.md:187:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 187,
        "line_end": 197,
        "snippet": "```"
      },
      {
        "id": "blocker:SKILL.md:66:system-reconnaissance",
        "category": "blocker",
        "severity": "low",
        "pattern": "System reconnaissance",
        "file": "SKILL.md",
        "line_start": 66,
        "line_end": 66,
        "snippet": "- **Avoid duplication**: Information should live in either SKILL.md or references files, not both. P"
      },
      {
        "id": "blocker:SKILL.md:104:system-reconnaissance",
        "category": "blocker",
        "severity": "low",
        "pattern": "System reconnaissance",
        "file": "SKILL.md",
        "line_start": 104,
        "line_end": 104,
        "snippet": "To avoid overwhelming users, avoid asking too many questions in a single message. Start with the mos"
      }
    ],
    "finding_verdicts": [
      {
        "id": "external_commands:scripts/init_skill.py:73:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "The flagged text is inside a generated Markdown template that lists example script filenames in inline code. It is not executed and is not passed to a shell."
      },
      {
        "id": "external_commands:scripts/init_skill.py:74:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "The flagged text is inside a generated Markdown template that lists example document-processing filenames. It is documentation text, not shell execution."
      },
      {
        "id": "external_commands:scripts/init_skill.py:84:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "The flagged text is inside a generated reference-document template and uses inline code formatting for filenames. No interpreter evaluates the backticks."
      },
      {
        "id": "filesystem:scripts/package_skill.py:14:python-archive-libraries",
        "verdict": "false_positive",
        "confidence": 0.88,
        "reason": "This line only imports Python zipfile for later packaging. An import alone does not read, extract, or execute archive contents."
      },
      {
        "id": "filesystem:scripts/package_skill.py:68:python-archive-libraries",
        "verdict": "confirmed",
        "severity": "medium",
        "confidence": 0.8,
        "reason": "This line creates a zip archive for a caller-provided skill path. The packager recursively writes local files into the archive, so it can disclose unintended contents."
      },
      {
        "id": "blocker:scripts/quick_validate.py:28:system-reconnaissance",
        "verdict": "false_positive",
        "confidence": 0.97,
        "reason": "This return reports invalid frontmatter in the user-provided skill file. It does not collect or expose host system information."
      },
      {
        "id": "blocker:scripts/quick_validate.py:65:system-reconnaissance",
        "verdict": "false_positive",
        "confidence": 0.95,
        "reason": "This exits with the validator status code after printing the validation result. It does not inspect host state beyond the user-provided skill path."
      },
      {
        "id": "external_commands:SKILL.md:29:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.99,
        "reason": "The range is a Markdown code fence showing the directory anatomy of a skill. It is documentation only and contains no shell substitution or executable command."
      },
      {
        "id": "external_commands:SKILL.md:40:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.99,
        "reason": "This is the closing fence for a Markdown directory diagram. It is not executable content and cannot invoke a shell."
      },
      {
        "id": "external_commands:SKILL.md:44:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.99,
        "reason": "The backticks mark YAML field names such as name and description. They are inline Markdown formatting, not command execution."
      },
      {
        "id": "external_commands:SKILL.md:48:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.97,
        "reason": "The backticks mark the scripts directory name in prose. This line describes a resource category and does not invoke anything."
      },
      {
        "id": "external_commands:SKILL.md:53:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.98,
        "reason": "The line names an example script path using inline code formatting. It is descriptive text and is not evaluated by a shell."
      },
      {
        "id": "external_commands:SKILL.md:57:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.98,
        "reason": "The backticks mark the references directory name in prose. There is no command or dynamic execution context."
      },
      {
        "id": "external_commands:SKILL.md:62:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.98,
        "reason": "The flagged text names reference-document examples with inline code formatting. These are documentation paths, not commands."
      },
      {
        "id": "external_commands:SKILL.md:68:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.98,
        "reason": "The backticks mark the assets directory name in a heading. It is Markdown formatting and contains no executable instruction."
      },
      {
        "id": "external_commands:SKILL.md:73:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.98,
        "reason": "The line lists asset filename examples using inline code. It is documentation for file organization and does not execute a command."
      },
      {
        "id": "external_commands:SKILL.md:115:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.98,
        "reason": "The backticks mark an example skill name in prose. No shell syntax or execution path is present."
      },
      {
        "id": "external_commands:SKILL.md:118:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "The line names a possible script path as a planning example. It does not run the script or build a shell command."
      },
      {
        "id": "external_commands:SKILL.md:120:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.98,
        "reason": "The backticks mark an example frontend skill name. This is inline Markdown formatting, not execution."
      },
      {
        "id": "external_commands:SKILL.md:123:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.98,
        "reason": "The line names an asset template directory using inline code. It describes bundled resources and does not invoke external commands."
      },
      {
        "id": "external_commands:SKILL.md:125:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.98,
        "reason": "The backticks mark an example skill name in prose. There is no command invocation or shell substitution."
      },
      {
        "id": "external_commands:SKILL.md:128:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.98,
        "reason": "The line names a reference document path using inline code. It is not executable content."
      },
      {
        "id": "external_commands:SKILL.md:138:ruby-shell-backtick-execution",
        "verdict": "confirmed",
        "severity": "medium",
        "confidence": 0.82,
        "reason": "The text instructs an agent to run the bundled init_skill.py script when creating a new skill. That script writes directories and files under user-provided inputs, so this is a real local command and filesystem-write risk."
      },
      {
        "id": "external_commands:SKILL.md:142:ruby-shell-backtick-execution",
        "verdict": "confirmed",
        "severity": "medium",
        "confidence": 0.84,
        "reason": "The fenced bash example shows executing scripts/init_skill.py with user-supplied skill name and output path. The target script is local and legitimate, but following it runs code that creates files."
      },
      {
        "id": "external_commands:SKILL.md:144:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.87,
        "reason": "This finding points at the closing code fence and descriptive bullets for the initializer example. The executable command is covered by the prior finding, so this backtick match is not a separate risk."
      },
      {
        "id": "external_commands:SKILL.md:150:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This line names generated directories in inline code. It describes files created by the initializer and does not execute a command."
      },
      {
        "id": "external_commands:SKILL.md:161:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "This line references resource directory names with inline code formatting. It is implementation guidance and contains no command invocation."
      },
      {
        "id": "external_commands:SKILL.md:163:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.95,
        "reason": "The line discusses deleting unneeded example files from generated skills. It does not include shell execution or host reconnaissance."
      },
      {
        "id": "external_commands:SKILL.md:179:ruby-shell-backtick-execution",
        "verdict": "confirmed",
        "severity": "medium",
        "confidence": 0.84,
        "reason": "The fenced bash example tells the agent to run scripts/package_skill.py on a selected skill folder. That bundled script creates a zip archive from local files, which is a real command and filesystem operation."
      },
      {
        "id": "external_commands:SKILL.md:181:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.88,
        "reason": "This finding points at the closing fence and prose between packaging examples. The actual packaging command is covered by the previous finding."
      },
      {
        "id": "external_commands:SKILL.md:185:ruby-shell-backtick-execution",
        "verdict": "confirmed",
        "severity": "medium",
        "confidence": 0.84,
        "reason": "The fenced bash example gives an alternate package_skill.py invocation with an output directory. Running it writes a zip archive to a user-selected location."
      },
      {
        "id": "external_commands:SKILL.md:187:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.88,
        "reason": "This finding points at the closing fence and descriptive validation bullets. The actual package command is covered by the preceding finding."
      },
      {
        "id": "blocker:SKILL.md:66:system-reconnaissance",
        "verdict": "false_positive",
        "confidence": 0.98,
        "reason": "This line advises keeping documentation concise and avoiding duplicated content. It does not collect, inspect, or transmit system information."
      },
      {
        "id": "blocker:SKILL.md:104:system-reconnaissance",
        "verdict": "false_positive",
        "confidence": 0.98,
        "reason": "This line advises limiting questions to users. It is conversation guidance and has no system reconnaissance behavior."
      }
    ],
    "semantic_findings": [
      {
        "title": "Path-Like Skill Names Can Escape the Target Directory",
        "description": "init_skill combines a resolved base path with the caller-supplied skill_name before creating files. Because skill_name is not validated, names containing path separators or traversal segments can create a skill directory outside the intended output folder.",
        "severity": "medium",
        "locations": [
          {
            "file": "scripts/init_skill.py",
            "line_start": 194,
            "line_end": 207
          },
          {
            "file": "scripts/init_skill.py",
            "line_start": 287,
            "line_end": 288
          }
        ],
        "confidence": 0.86,
        "confidence_reasoning": "The script uses argv values directly and does not enforce the printed hyphen-case naming rules before building the output path. The impact depends on the caller supplying a path-like skill name."
      },
      {
        "title": "Unfiltered Recursive Packaging May Include Secrets",
        "description": "package_skill walks every file under the selected skill directory and adds it to the zip archive. Without exclusions for hidden files, credentials, caches, VCS folders, or symlink targets, authors can accidentally publish sensitive local files.",
        "severity": "medium",
        "locations": [
          {
            "file": "scripts/package_skill.py",
            "line_start": 70,
            "line_end": 74
          }
        ],
        "confidence": 0.84,
        "confidence_reasoning": "The code directly recurses through the selected skill directory and writes each file to the archive. The exposure depends on files or symlinks present in the folder, so confidence is high but not absolute."
      }
    ],
    "critical_findings": [],
    "high_findings": [],
    "medium_findings": [
      {
        "title": "Python archive libraries",
        "description": "with zipfile.ZipFile(zip_filename, 'w', zipfile.ZIP_DEFLATED) as zipf:",
        "locations": [
          {
            "file": "scripts/package_skill.py",
            "line_start": 68,
            "line_end": 68
          }
        ],
        "review_kind": "capability",
        "source_category": "filesystem",
        "source_severity": "medium",
        "confidence": 0.8,
        "confidence_reasoning": "This line creates a zip archive for a caller-provided skill path. The packager recursively writes local files into the archive, so it can disclose unintended contents."
      },
      {
        "title": "Ruby/shell backtick execution",
        "description": "When creating a new skill from scratch, always run the `init_skill.py` script. The script convenient",
        "locations": [
          {
            "file": "SKILL.md",
            "line_start": 138,
            "line_end": 142
          }
        ],
        "review_kind": "capability",
        "source_category": "external_commands",
        "source_severity": "medium",
        "confidence": 0.82,
        "confidence_reasoning": "The text instructs an agent to run the bundled init_skill.py script when creating a new skill. That script writes directories and files under user-provided inputs, so this is a real local command and filesystem-write risk."
      },
      {
        "title": "Ruby/shell backtick execution",
        "description": "```bash",
        "locations": [
          {
            "file": "SKILL.md",
            "line_start": 142,
            "line_end": 144
          }
        ],
        "review_kind": "capability",
        "source_category": "external_commands",
        "source_severity": "medium",
        "confidence": 0.84,
        "confidence_reasoning": "The fenced bash example shows executing scripts/init_skill.py with user-supplied skill name and output path. The target script is local and legitimate, but following it runs code that creates files."
      },
      {
        "title": "Ruby/shell backtick execution",
        "description": "```bash",
        "locations": [
          {
            "file": "SKILL.md",
            "line_start": 179,
            "line_end": 181
          }
        ],
        "review_kind": "capability",
        "source_category": "external_commands",
        "source_severity": "medium",
        "confidence": 0.84,
        "confidence_reasoning": "The fenced bash example tells the agent to run scripts/package_skill.py on a selected skill folder. That bundled script creates a zip archive from local files, which is a real command and filesystem operation."
      },
      {
        "title": "Ruby/shell backtick execution",
        "description": "```bash",
        "locations": [
          {
            "file": "SKILL.md",
            "line_start": 185,
            "line_end": 187
          }
        ],
        "review_kind": "capability",
        "source_category": "external_commands",
        "source_severity": "medium",
        "confidence": 0.84,
        "confidence_reasoning": "The fenced bash example gives an alternate package_skill.py invocation with an output directory. Running it writes a zip archive to a user-selected location."
      },
      {
        "title": "Path-Like Skill Names Can Escape the Target Directory",
        "description": "init_skill combines a resolved base path with the caller-supplied skill_name before creating files. Because skill_name is not validated, names containing path separators or traversal segments can create a skill directory outside the intended output folder.",
        "locations": [
          {
            "file": "scripts/init_skill.py",
            "line_start": 194,
            "line_end": 207
          },
          {
            "file": "scripts/init_skill.py",
            "line_start": 287,
            "line_end": 288
          }
        ],
        "review_kind": "security",
        "source_category": "semantic",
        "source_severity": "medium",
        "confidence": 0.86,
        "confidence_reasoning": "The script uses argv values directly and does not enforce the printed hyphen-case naming rules before building the output path. The impact depends on the caller supplying a path-like skill name."
      },
      {
        "title": "Unfiltered Recursive Packaging May Include Secrets",
        "description": "package_skill walks every file under the selected skill directory and adds it to the zip archive. Without exclusions for hidden files, credentials, caches, VCS folders, or symlink targets, authors can accidentally publish sensitive local files.",
        "locations": [
          {
            "file": "scripts/package_skill.py",
            "line_start": 70,
            "line_end": 74
          }
        ],
        "review_kind": "security",
        "source_category": "semantic",
        "source_severity": "medium",
        "confidence": 0.84,
        "confidence_reasoning": "The code directly recurses through the selected skill directory and writes each file to the archive. The exposure depends on files or symlinks present in the folder, so confidence is high but not absolute."
      }
    ],
    "low_findings": [],
    "dangerous_patterns": [],
    "files_scanned": 5,
    "total_lines": 892,
    "audit_model": "codex",
    "audited_at": "2026-07-05T16:11:28.099Z",
    "agent_auto_install_policy": "allowed",
    "manual_install_policy": "allowed"
  },
  "content": {
    "user_title": "Create Reliable AI Skills",
    "value_statement": "Building reusable AI skills is hard when structure, metadata, and packaging rules are unclear. This skill guides Claude, Codex, and Claude Code through planning, scaffolding, editing, validation, and packaging.",
    "seo_keywords": [
      "Claude",
      "Codex",
      "Claude Code",
      "AI skill creation",
      "skill marketplace",
      "SKILL.md guidance",
      "Claude skills",
      "Codex skills",
      "skill packaging",
      "skill validation"
    ],
    "actual_capabilities": [
      "Explains the required SKILL.md structure, YAML metadata, and optional bundled resources.",
      "Guides authors through gathering concrete examples before writing a skill.",
      "Helps choose when to use scripts, references, and assets for reusable work.",
      "Provides an initializer script that creates a template skill directory.",
      "Provides validation and packaging scripts for distributable skill zip files."
    ],
    "limitations": [
      "It does not publish skills to a marketplace or remote registry.",
      "It only performs basic validation and does not replace a security audit.",
      "Generated templates still need domain content, testing, and cleanup before release.",
      "Authors must review skill names, paths, and package contents before running helper scripts."
    ],
    "use_cases": [
      {
        "title": "Create an Internal Automation Skill",
        "description": "Plan and scaffold a skill that captures repeatable team workflows, helper scripts, and reference documents.",
        "target_user": "Developer"
      },
      {
        "title": "Document a Team Workflow",
        "description": "Convert a known process into clear skill instructions with examples, resource choices, and validation steps.",
        "target_user": "Team lead"
      },
      {
        "title": "Prepare a Community Skill Package",
        "description": "Review structure, remove placeholder files, validate metadata, and package the skill for distribution.",
        "target_user": "Marketplace author"
      }
    ],
    "prompt_templates": [
      {
        "title": "Plan a Simple Skill",
        "scenario": "Beginner planning from an idea",
        "prompt": "I want to create a skill for [task]. Ask the key questions, then propose a concise SKILL.md outline and needed resources."
      },
      {
        "title": "Scaffold a New Skill",
        "scenario": "Create the initial folder structure",
        "prompt": "Use the skill-creator workflow to initialize a new skill named [skill-name] in [workspace path], then explain what files I should customize."
      },
      {
        "title": "Improve an Existing Skill",
        "scenario": "Refine quality and structure",
        "prompt": "Review my existing skill files for metadata quality, resource organization, missing examples, and leftover template content."
      },
      {
        "title": "Prepare for Release",
        "scenario": "Advanced marketplace readiness review",
        "prompt": "Evaluate this skill for release readiness. Check validation issues, packaging risks, unnecessary files, and whether the instructions support repeated use."
      }
    ],
    "output_examples": [
      {
        "input": "Create a skill for turning meeting notes into action plans.",
        "output": [
          "Recommended structure: workflow-based skill with intake, extraction, prioritization, and review sections.",
          "Reusable resources: a reference checklist for action item quality and optional templates for final summaries.",
          "Next step: gather two real note samples before finalizing the instructions."
        ]
      },
      {
        "input": "Review my skill before packaging.",
        "output": [
          "Metadata is specific enough to trigger on the intended task.",
          "The references folder is useful, but one placeholder asset should be removed.",
          "Package only after checking for private files, symlinks, and generated archives."
        ]
      },
      {
        "input": "Help me decide whether this skill needs a script.",
        "output": [
          "A script is useful because the task repeats deterministic file processing steps.",
          "Keep business rules in SKILL.md and put long API details in references.",
          "Document script inputs, outputs, and failure cases before release."
        ]
      }
    ],
    "best_practices": [
      "Start with real user examples before writing the final skill instructions.",
      "Keep SKILL.md concise and move detailed reference material into references.",
      "Use hyphen-case skill names and inspect package contents before distribution."
    ],
    "anti_patterns": [
      "Publishing a skill that still contains TODO placeholders or generic template text.",
      "Bundling large reference files without clear loading guidance or search patterns.",
      "Using path-like skill names or unchecked folders when running helper scripts."
    ],
    "faq": [
      {
        "question": "What does this skill help create?",
        "answer": "It helps create reusable AI skills with clear instructions, metadata, optional resources, validation, and packaging."
      },
      {
        "question": "Can it work with Claude, Codex, and Claude Code?",
        "answer": "Yes. The marketplace metadata lists support for Claude, Codex, and Claude Code."
      },
      {
        "question": "Does it generate a complete finished skill automatically?",
        "answer": "No. It creates structure and guidance, but authors must add domain content, examples, and tests."
      },
      {
        "question": "Does it include validation?",
        "answer": "Yes. It includes a basic validator for frontmatter, naming, description format, and required files."
      },
      {
        "question": "What files can the initializer create?",
        "answer": "It can create SKILL.md plus example scripts, references, and assets folders for customization."
      },
      {
        "question": "Should authors remove generated example files?",
        "answer": "Yes. Authors should delete unused examples, placeholders, private files, symlinks, and generated archives before packaging."
      }
    ]
  },
  "file_structure": [
    {
      "name": "LICENSE.txt",
      "type": "file",
      "path": "LICENSE.txt",
      "lines": 202
    },
    {
      "name": "scripts",
      "type": "dir",
      "path": "scripts",
      "children": [
        {
          "name": "init_skill.py",
          "type": "file",
          "path": "scripts/init_skill.py",
          "lines": 304
        },
        {
          "name": "package_skill.py",
          "type": "file",
          "path": "scripts/package_skill.py",
          "lines": 111
        },
        {
          "name": "quick_validate.py",
          "type": "file",
          "path": "scripts/quick_validate.py",
          "lines": 65
        }
      ]
    },
    {
      "name": "SKILL.md",
      "type": "file",
      "path": "SKILL.md",
      "lines": 210
    }
  ]
}
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 323kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 323kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 323kUpdated Claude CodeCodexVerifiedInstall