Mine coding agent logs (Codex/Cursor/session histories and similar telemetry) to discover high-value candidate skills, then draft structured skill creation/reuse recommendations.
Ask your Agent
Use this Skill: https://skilld.dev/gh/vincentkoc/dotskills/technical-skill-finder
This session only. Nothing lands on disk.
≈124 tokens on demand. Your agent reads this file only when SKILL.md points to it.
Score each candidate 0-5 in each dimension:
frequency: repeated occurrences across distinct sessionsimpact: time-to-resolution / incident severityactionability: can the pattern be handled by deterministic checks/workflowstoolability: enough local tooling commands to automatenovelty: not already covered by existing skill coverageconfidence = average score / 5.
Prioritize candidates with:
Source: SKILL.md on GitHub
This skill mines sensitive local agent interaction histories and telemetry to suggest new automation. It lacks safeguards against Indirect Prompt Injection, meaning malicious instructions embedded in historical logs could trick the agent into recommending harmful commands or exposing private conversation data. Additionally, it reads dotfiles and log directories that often contain sensitive credentials and proprietary code.
No alerts
Risk: LOW · No issues
4/4 files flagged
Signed by skilld at e0b7247. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.
Last checked against GitHub last week.
{
"source": "https://github.com/vincentkoc/dotskills"
}