resourceful-problem-solving — Claude Code skill: setup and features
This skill gives coding agents a fallback plan for requests they cannot handle directly: look for an existing skill, find a command-line utility, or write a script.
npx skills add swarmclawai/swarmclaw --skill resourceful-problem-solving -a claude-codeWhat resourceful-problem-solving does
resourceful-problem-solving defines a sequence for handling tasks when no dedicated tool is available. The agent first searches for a relevant skill with use_skill, then checks for a suitable command-line utility. If neither route works, it creates a small solution in Python, Node.js, or Bash.
The guidance names tools for specific jobs. PDF generation can use reportlab or fpdf2, image work can use Pillow, spreadsheet handling can use openpyxl, and data conversion can use jq. For extracting website content, it suggests curl with response parsing or Python packages such as requests and beautifulsoup4. Dependencies can be added with pip install <package> or npm install <package>.
The skill also recommends offering to preserve useful scripts through extension_creator when the same task is likely to recur. Agents are expected to inspect their output before claiming completion. That check matters: finding a tool or writing code does not, by itself, establish that the task succeeded. The skill is designed to stay active and supports Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, and OpenCode.
What it helps with
- Find an existing skill for a task without a dedicated tool
- Check for a CLI utility and install the required package
- Create a script that generates PDFs or edits images
- Convert spreadsheet data with a suitable command or script
- Turn a recurring script into a reusable extension
Example prompts for the agent
- Find a skill that can edit my PDF; if none fits, look for a CLI tool or write a script.
- Convert this spreadsheet to the required format, checking available tools before creating a custom solution.
- Write a script to collect website data, verify its output, and offer to save it for future use.
How to install resourceful-problem-solving
1Skills CLI (Vercel)
npx skills add swarmclawai/swarmclaw --skill resourceful-problem-solving -a claude-code2Manually: put the files in place
~/.claude/skills/resourceful-problem-solving/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/resourceful-problem-solving/
1Skills CLI (Vercel)
npx skills add swarmclawai/swarmclaw --skill resourceful-problem-solving -a codex2Manually: put the files in place
.agents/skills/resourceful-problem-solving/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add swarmclawai/swarmclaw --skill resourceful-problem-solving -a cursor2Manually: put the files in place
.cursor/skills/resourceful-problem-solving/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install swarmclawai/swarmclaw resourceful-problem-solving2Manually: put the files in place
.github/skills/resourceful-problem-solving/SKILL.md1Skills CLI (Vercel)
npx skills add swarmclawai/swarmclaw --skill resourceful-problem-solving -a gemini-cli2Platform CLI
gemini skills install https://github.com/swarmclawai/swarmclaw --path skills/resourceful-problem-solvingManually: put the files in place
.devin/skills/resourceful-problem-solving/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add swarmclawai/swarmclaw --skill resourceful-problem-solving -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about resourceful-problem-solving
How do I install resourceful-problem-solving in Claude Code?
Run npx skills add swarmclawai/swarmclaw --skill resourceful-problem-solving -a claude-code. The skill file is located at ~/.claude/skills/resourceful-problem-solving/SKILL.md.
What does resourceful-problem-solving do?
It gives an agent a fallback workflow for tasks that lack a dedicated tool: search for an existing skill, check command-line options, and write a script when needed. It also calls for output verification and offers a way to preserve reusable scripts.
Which coding platforms support resourceful-problem-solving?
The skill supports Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, and OpenCode. Each platform has its own installation command or documented skill location.
Source text
SKILL.md · MIT
--- name: resourceful-problem-solving description: Always-on guidance for solving tasks resourcefully. Teaches agents to escalate through skills, CLI tools, and custom scripts instead of refusing. Applies to any request where the agent lacks a dedicated tool. always: true --- # Resourceful Problem Solving When asked to do something and no dedicated tool exists, follow this escalation pattern: ## 1. Check Skills Use `use_skill` to search for a matching skill. Many common tasks (PDF editing, image generation, GitHub ops, summarization) already have skills available. ## 2. Search for a CLI Tool If no skill matches, check whether a CLI tool exists: ```bash which <tool> # already installed? brew search <tool> # macOS pip search <tool> # Python ecosystem npm search <tool> # Node ecosystem ``` If a suitable tool exists, install it and use it. ## 3. Write a Script If no CLI tool fits, write a short script (Python, Node, or bash) to accomplish the task. Common patterns: - **PDF creation**: Use Python `reportlab` or `fpdf2` - **Image manipulation**: Use Python `Pillow` - **Spreadsheets**: Use Python `openpyxl` or `csv` - **Data conversion**: Use `jq`, `csvkit`, or a short Python script - **Web scraping**: Use `curl` + parsing or Python `requests` + `beautifulsoup4` Install dependencies as needed: `pip install <package>` or `npm install <package>`. ## 4. Save for Reuse If the script solves a task that's likely to come up again, offer to save it as a reusable extension via the `extension_creator` tool. ## Rules - Never refuse a task when shell tools are available. Attempt a code-based solution first. - Only report inability after genuinely trying and failing. - Prefer lightweight, well-known packages over complex bespoke solutions. - Always verify the output before reporting success.
© swarmclawai · MIT · swarmclawai/swarmclaw
