Skip to content

aiskillstore/marketplace / skills/0xmsc/create-skill

create-skill

Verified

Guide for creating new Agent Skills. Use this skill when you need to create a new skill.

FileKindSize
SKILL.mdskill md1.2 KB
references/specification.mdmarkdown6.4 KB
references/what-are-skills.mdmarkdown2.6 KB
skill-report.jsonconfig16.4 KB

skill-report.json

View on GitHub
{
  "schema_version": "2.0",
  "meta": {
    "generated_at": "2026-07-06T00:59:59.327Z",
    "slug": "0xmsc-create-skill",
    "source_url": "https://github.com/0xmsc/coding_assistant/tree/master/src/coding_assistant/skills/create-skill",
    "source_ref": "master",
    "model": "codex",
    "analysis_version": "3.0.0",
    "source_type": "community",
    "content_hash": "c48547ab10095b74e2f617672312ecb34d367d59dc3f1f010c04891788583224",
    "tree_hash": "c27803eb592e0622c5caa9c8f2f9ef424b8af0381feebcc61972d1c4409a8fd8",
    "provenance": {
      "model_requested": "gpt-5.5:high",
      "model_effective": "codex:gpt-5.5:high",
      "fallback_chain": [
        {
          "agent": "codex",
          "model": "gpt-5.5",
          "effort": "high",
          "outcome": "success"
        }
      ],
      "analysis_version": "3.0.0"
    }
  },
  "skill": {
    "name": "create-skill",
    "description": "Helps authors create Agent Skills with the required folder structure, frontmatter, references, and validation steps.",
    "summary": "Create well-structured Agent Skills with concise instructions and supporting resources.",
    "icon": "πŸ“¦",
    "version": "1.0.0",
    "author": "0xmsc",
    "license": "MIT",
    "tags": [
      "agent-skills",
      "skill-authoring",
      "documentation",
      "workflow-design"
    ],
    "supported_tools": [
      "claude",
      "codex",
      "claude-code"
    ],
    "risk_factors": [
      "external_commands"
    ],
    "category": "documentation"
  },
  "security_audit": {
    "risk_level": "safe",
    "is_blocked": false,
    "safe_to_publish": true,
    "analysis_status": "ok",
    "summary": "Static analysis flagged Markdown examples and inline code formatting as reconnaissance or shell execution. Review found these are documentation-only authoring instructions, with no prompt injection, data exfiltration, hidden automation, or malicious intent.",
    "remediation": [],
    "risk_factor_evidence": [
      {
        "factor": "external_commands",
        "evidence": [
          {
            "file": "SKILL.md",
            "line_start": 9,
            "line_end": 15
          },
          {
            "file": "SKILL.md",
            "line_start": 15,
            "line_end": 18
          },
          {
            "file": "SKILL.md",
            "line_start": 18,
            "line_end": 18
          },
          {
            "file": "SKILL.md",
            "line_start": 21,
            "line_end": 21
          },
          {
            "file": "SKILL.md",
            "line_start": 24,
            "line_end": 24
          },
          {
            "file": "SKILL.md",
            "line_start": 25,
            "line_end": 27
          }
        ]
      }
    ],
    "static_findings": [
      {
        "id": "blocker:references/specification.md:65:system-reconnaissance",
        "category": "blocker",
        "severity": "low",
        "pattern": "System reconnaissance",
        "file": "references/specification.md",
        "line_start": 65,
        "line_end": 65,
        "snippet": "Valid examples:"
      },
      {
        "id": "blocker:references/specification.md:76:system-reconnaissance",
        "category": "blocker",
        "severity": "low",
        "pattern": "System reconnaissance",
        "file": "references/specification.md",
        "line_start": 76,
        "line_end": 76,
        "snippet": "Invalid examples:"
      },
      {
        "id": "blocker:references/specification.md:139:system-reconnaissance",
        "category": "blocker",
        "severity": "low",
        "pattern": "System reconnaissance",
        "file": "references/specification.md",
        "line_start": 139,
        "line_end": 139,
        "snippet": "- We recommend making your key names reasonably unique to avoid accidental conflicts"
      },
      {
        "id": "blocker:references/specification.md:218:system-reconnaissance",
        "category": "blocker",
        "severity": "low",
        "pattern": "System reconnaissance",
        "file": "references/specification.md",
        "line_start": 218,
        "line_end": 218,
        "snippet": "Keep file references one level deep from `SKILL.md`. Avoid deeply nested reference chains."
      },
      {
        "id": "blocker:references/specification.md:222:system-reconnaissance",
        "category": "blocker",
        "severity": "low",
        "pattern": "System reconnaissance",
        "file": "references/specification.md",
        "line_start": 222,
        "line_end": 222,
        "snippet": "Verify that your `SKILL.md` frontmatter is valid and follows all naming conventions as defined in th"
      },
      {
        "id": "external_commands:SKILL.md:9:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 9,
        "line_end": 15,
        "snippet": "```"
      },
      {
        "id": "external_commands:SKILL.md:15:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 15,
        "line_end": 18,
        "snippet": "```"
      },
      {
        "id": "external_commands:SKILL.md:18:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 18,
        "line_end": 18,
        "snippet": "- **Frontmatter**: Must include `name` (lowercase-alphanumeric) and `description`."
      },
      {
        "id": "external_commands:SKILL.md:21:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 21,
        "line_end": 21,
        "snippet": "- **Resources**: Move detailed docs to `references/` and code to `scripts/`."
      },
      {
        "id": "external_commands:SKILL.md:24:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 24,
        "line_end": 24,
        "snippet": "1. **Directory**: `mkdir <name>`. Create `scripts/`, `references/`, or `assets/` only if you actuall"
      },
      {
        "id": "external_commands:SKILL.md:25:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 25,
        "line_end": 27,
        "snippet": "2. **Metadata**: Write the `SKILL.md` frontmatter first."
      }
    ],
    "finding_verdicts": [
      {
        "id": "blocker:references/specification.md:65:system-reconnaissance",
        "verdict": "false_positive",
        "confidence": 0.97,
        "reason": "Line 65 introduces valid frontmatter name examples in a specification document. It does not request system inspection or reconnaissance."
      },
      {
        "id": "blocker:references/specification.md:76:system-reconnaissance",
        "verdict": "false_positive",
        "confidence": 0.97,
        "reason": "Line 76 introduces invalid naming examples for skill authors. The text is validation guidance, not an instruction to enumerate system details."
      },
      {
        "id": "blocker:references/specification.md:139:system-reconnaissance",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "The line advises unique metadata keys to avoid conflicts. It does not ask the agent to discover host, user, environment, or network information."
      },
      {
        "id": "blocker:references/specification.md:218:system-reconnaissance",
        "verdict": "false_positive",
        "confidence": 0.98,
        "reason": "This is authoring guidance about shallow file references from SKILL.md. It limits reference depth and does not direct filesystem reconnaissance."
      },
      {
        "id": "blocker:references/specification.md:222:system-reconnaissance",
        "verdict": "false_positive",
        "confidence": 0.98,
        "reason": "The line asks authors to validate frontmatter naming conventions. That is local document validation, not system reconnaissance."
      },
      {
        "id": "external_commands:SKILL.md:9:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "The finding points to the opening fence of a Markdown directory tree. There is no Ruby backtick execution or runnable command at this location."
      },
      {
        "id": "external_commands:SKILL.md:15:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.96,
        "reason": "The finding points to the closing fence of a Markdown directory tree. It is static documentation text and does not execute a shell command."
      },
      {
        "id": "external_commands:SKILL.md:18:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.94,
        "reason": "The backticks mark literal field names in Markdown frontmatter guidance. They do not invoke Ruby or a shell."
      },
      {
        "id": "external_commands:SKILL.md:21:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.94,
        "reason": "The backticks identify folder names for skill resources. This is documentation of expected structure, not command execution."
      },
      {
        "id": "external_commands:SKILL.md:24:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.86,
        "reason": "The inline mkdir example is authoring guidance for creating a skill folder. It is not hidden automation, and the surrounding requirements constrain the name format."
      },
      {
        "id": "external_commands:SKILL.md:25:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.94,
        "reason": "The backticks mark the literal SKILL.md filename in a process step. There is no executable script, shell interpolation, or Ruby code."
      }
    ],
    "semantic_findings": [],
    "critical_findings": [],
    "high_findings": [],
    "medium_findings": [],
    "low_findings": [],
    "dangerous_patterns": [],
    "files_scanned": 3,
    "total_lines": 325,
    "audit_model": "codex",
    "audited_at": "2026-07-06T00:59:59.327Z",
    "agent_auto_install_policy": "allowed",
    "manual_install_policy": "allowed"
  },
  "content": {
    "user_title": "Create Agent Skills Faster",
    "value_statement": "Writing Agent Skills requires clear structure and valid metadata. This skill guides authors through folders, frontmatter, references, and validation.",
    "seo_keywords": [
      "Claude",
      "Codex",
      "Claude Code",
      "Agent Skills",
      "create skill",
      "skill authoring",
      "SKILL.md",
      "frontmatter",
      "AI workflow documentation",
      "skill validation"
    ],
    "actual_capabilities": [
      "Explains the required SKILL.md frontmatter fields for a new skill.",
      "Shows the standard folder layout for scripts, references, and assets.",
      "Provides a short process for creating a skill directory and metadata.",
      "Links to a full Agent Skills format specification.",
      "Links to design guidance for progressive disclosure and skill structure."
    ],
    "limitations": [
      "Does not include automation scripts for generating files.",
      "Does not validate a completed skill with a bundled checker.",
      "Does not provide domain-specific examples beyond general skill authoring.",
      "Does not manage publishing, packaging, or marketplace submission."
    ],
    "use_cases": [
      {
        "title": "Draft a first skill",
        "description": "Create the initial folder, SKILL.md metadata, and concise body instructions for a new Agent Skill.",
        "target_user": "Individual developer"
      },
      {
        "title": "Standardize team skill docs",
        "description": "Use one structure for naming, references, scripts, and validation across team-authored skills.",
        "target_user": "Engineering team lead"
      },
      {
        "title": "Review skill structure",
        "description": "Check whether a proposed skill uses valid frontmatter and a clear resource layout.",
        "target_user": "Marketplace reviewer"
      }
    ],
    "prompt_templates": [
      {
        "title": "Create a basic skill",
        "scenario": "Beginner authoring task",
        "prompt": "Create a new Agent Skill named <skill-name>. Include valid SKILL.md frontmatter and a short instruction body."
      },
      {
        "title": "Plan skill resources",
        "scenario": "Adding supporting files",
        "prompt": "Plan the folder structure for a skill that needs scripts and reference documents. Explain what belongs in each folder."
      },
      {
        "title": "Improve an existing skill",
        "scenario": "Quality pass",
        "prompt": "Review this SKILL.md for metadata quality, concise instructions, and appropriate references. Suggest precise improvements."
      },
      {
        "title": "Design a production-ready skill",
        "scenario": "Advanced authoring workflow",
        "prompt": "Help me design a complete Agent Skill for <workflow>. Define activation wording, resource files, validation checks, and edge cases."
      }
    ],
    "output_examples": [
      {
        "input": "Create a skill for reviewing pull requests.",
        "output": "A skill plan with a lowercase name, a clear description, a short SKILL.md outline, and suggested reference files."
      },
      {
        "input": "Check whether my skill folder is organized correctly.",
        "output": [
          "Confirms whether SKILL.md is present and named correctly.",
          "Identifies optional folders that have a clear purpose.",
          "Flags overly deep references or missing metadata."
        ]
      },
      {
        "input": "Turn my long skill instructions into a cleaner structure.",
        "output": "A concise main instruction flow with detailed material moved into focused reference documents."
      }
    ],
    "best_practices": [
      "Keep SKILL.md concise and move detailed material into focused reference files.",
      "Write descriptions that state both what the skill does and when to use it.",
      "Use only optional folders that contain real scripts, references, or assets."
    ],
    "anti_patterns": [
      "Do not create empty scripts, references, or assets folders.",
      "Do not use vague descriptions that hide the skill trigger conditions.",
      "Do not bury core instructions behind many nested reference files."
    ],
    "faq": [
      {
        "question": "What does this skill help me create?",
        "answer": "It helps create Agent Skills that use SKILL.md metadata, clear instructions, and optional supporting resources."
      },
      {
        "question": "Does it generate files automatically?",
        "answer": "No. It provides authoring guidance and process steps, but it does not include a generation script."
      },
      {
        "question": "Can I use it with Claude Code?",
        "answer": "Yes. The guidance is compatible with Claude Code, Codex, and Claude workflows that support Agent Skills."
      },
      {
        "question": "What should the description include?",
        "answer": "The description should explain what the skill does and when an agent should activate it."
      },
      {
        "question": "When should I add references?",
        "answer": "Add references when detailed material would make SKILL.md too long or hard to scan."
      },
      {
        "question": "Does the skill cover marketplace publishing?",
        "answer": "No. It focuses on authoring structure and validation, not marketplace packaging or submission."
      }
    ]
  },
  "file_structure": [
    {
      "name": "references",
      "type": "dir",
      "path": "references",
      "children": [
        {
          "name": "specification.md",
          "type": "file",
          "path": "references/specification.md",
          "lines": 223
        },
        {
          "name": "what-are-skills.md",
          "type": "file",
          "path": "references/what-are-skills.md",
          "lines": 70
        }
      ]
    },
    {
      "name": "SKILL.md",
      "type": "file",
      "path": "SKILL.md",
      "lines": 32
    }
  ]
}
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 β€” 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" β€” 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: 322kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 322kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself β€” your human partner chose inline execution, or no subagent tool is availableStars: 322kUpdated Claude CodeCodexVerifiedInstall