agents/log-tool-failure/SKILL.md
If any tool call fails, log it IMMEDIATELY with this skill
npx skillsauth add sanand0/scripts log-tool-failureInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Append each tool failure as single-line objects into ~/Documents/data/agents/tool-failure.jsonl. Schema:
{
"agent": "codex", // e.g. copilot, claude, ...
"path": "/home/sanand/...", // project directory
"session": "...", // session ID
"summary": "...", // crisp 1-sentence summary of what you tried, what failed, AND **WHY**
"request": {}, // complete tool call as JSON
"response": {} // complete tool response as JSON, including error code, message, etc.
}
testing
If the task had failures or was inefficient or complex, ALWAYS run this post-mortem.
data-ai
Brainstorm creative, novel, non-obvious ideas or options for an open-ended problem and narrow them to a recommendation. Not for writing prose, research, explanation, summaries, or questions with one right answer.
documentation
To write in Anand's style in blog posts, talk summaries, interview questions, emails, ...
documentation
Write data findings as a compelling narrative story, Malcolm Gladwell prose, NYT graphics-team visuals, engaging & memorable even for a non-technical audience.