Run an autonomous overnight campaign to find, prove, fix, and document CI flakes. Use when the user asks to stabilize CI through repeated full-pipeline runs and leave an evidence-backed draft PR.
Scanned 9/5/2026
Install to Claude Code
Skills Directorynpx -y skills add juspay/kolu --skill overnight-ci-flakes --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Overnight Ci Flakes?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/juspay-overnight-ci-flakes-236db0ea)More formats (shields.io, HTML) on the badges page.
---
name: overnight-ci-flakes
description: Run an autonomous overnight campaign to find, prove, fix, and document CI flakes. Use when the user asks to stabilize CI through repeated full-pipeline runs and leave an evidence-backed draft PR.
---
# Overnight CI flakes
- Ask for the exact macOS and Linux machine hosts and wait; never infer or reuse them.
- Open a draft PR and keep a concise `docs/ci-flakes.md` current with every run.
- Run `/ci` on both hosts without step retries until the current PR head passes 5 consecutive runs.
- For every failure, record evidence, proven root cause, proposed or applied fix, and present state; make every repository-addressable fix, reset the count, and repeat the 5-run loop. Claim no more specific a root cause than the failures prove, and scope the fix to that proven mechanism (arXiv:2301.12987).
- Only after 5 consecutive green runs, run `/be-review`.
- Run `/ci` 5 more consecutive times without step retries; if any run fails, return to the failure-and-fix loop above.
- Finish only when the final 5 runs and GitHub CI are green on the current PR head.
Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.
No comments yet. Be the first to comment!
Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...
Ultra-compressed communication mode. Cuts token usage ~75% by speaking like caveman while keeping full technical accuracy. Supports intensity levels: lite, full (default), ultra, wenyan-lite, wenyan-full, wenyan-ultra. Use when user says "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", or invokes /caveman. Also auto-triggers when token efficiency is requested.
Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.
**Complete production-ready guide for Google Gemini embeddings API** This skill provides comprehensive coverage of the `gemini-embedding-001` model for generating text embeddings, including SDK usage, REST API patterns, batch processing, RAG integration with Cloudflare Vectorize, and advanced use cases like semantic search and document clustering. ---
Recovers prior coding-agent session context by running `catchup <agent> --since-compact`, which extracts a clean summary of a previous Codex, Claude Code, Antigravity, OpenCode, or Pi Agent session. Use when the user says "catch up", "what did the last session do", "get me up to speed", "I switched agents", or asks to recover/summarize a previous session before continuing. Do NOT use for the current conversation, git history, or any non-agent log.
