New: 98,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

CI Red Sweep

by diegosouzapw in diegosouzapw/OmniRoute

Repairs confirmed CI failures on the base branch first, then prepares each affected PR one at a time with exact-commit evidence, never merging or weakening tests.

MITAuto-check passedDevOps & Cloud

Install CI Red Sweep

skills CLI
$ npx skills add diegosouzapw/OmniRoute --skill sweep-reds -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install diegosouzapw/OmniRoute sweep-reds --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/diegosouzapw/OmniRoute.git skills-src && mkdir -p .claude/skills && cp -r skills-src/config/quality/skill-templates/sweep-reds .claude/skills/sweep-reds && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
sweep-reds
GitHub stars
74k
Token cost
~868 tokens
SKILL.md length
448 words
Files
1
Skills in repo
50
Repo updated
First seen
Licence
MIT

At a glance

Repairs confirmed CI failures on the base branch first, then prepares each affected PR one at a time with exact-commit evidence, never merging or weakening tests.

  • Works in 4 steps: Resolve live main/release refs and… → Follow the shared exact-SHA evidence… → Reproduce each signature on an isolated… → …
  • Restoring a red main branch before reviewing other PRs
  • SKILL.md covers Boundaries, A. Base health, B. PR queue and Report
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

A disciplined routine for restoring a red CI queue. Part A fixes the base: it resolves live refs read-only, reproduces each failure signature on an isolated pinned base, opens one branch or worktree per independently reviewable repair with red-to-green proof, and hands the fix PR to its owner. The base is not called repaired until the fix is integrated and the new exact base passes.

Part B works the PR queue one PR at a time, sorting required failures, advisory failures, pending, stale, cancelled and approval-needed cases separately. Hard boundaries apply: no merging, closing PRs, changing release-freeze state, weakening assertions, raising baselines or editing CI under ordinary sweep authority, and missing access counts as a hold rather than a pass. It reads two shared files, base-green.md and validation-gate.md, first.

When your agent uses it

  • Restoring a red main branch before reviewing other PRs
  • Triaging which open PRs fail for the same base reason
  • Preparing a fix PR with red-to-green evidence
  • Sweeping a PR queue after a CI incident

Example prompts

  • “Sweep reds on the current queue, starting with the base branch.”
  • “Restore the queue: find why main is failing and prepare the fixes.”
  • “Which open PRs fail only because of the broken base? Classify them.”

Requirements

  • Read access to the repository's CI results and open PRs
  • The shared base-green.md and validation-gate.md files

Workflow steps

4 steps, taken from the first numbered list in SKILL.md.

  1. Resolve live main/release refs and freeze state read-only. Confirm the intended
  2. Follow the shared exact-SHA evidence protocol. Closed/missing incidents mean
  3. Reproduce each signature on an isolated pinned base. Keep setup/install failures
  4. Prepare/babysit the fix PR under the authorized scope, then hand it to the owner.

What it can do on your machine

Read from SKILL.md and the folder at commit f8a8131. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

CI Red Sweep loads about 868 tokens when it runs. Until then it costs about 65 tokens; SKILL.md has 448 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~65
When it runs · the whole SKILL.md, loaded when a task matches
~868

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from diegosouzapw/OmniRoute at commit f8a8131, republished under its MIT licence (© diegosouzapw). 448 words, ~868 tokens.

Download SKILL.mdSave it as .claude/skills/sweep-reds/SKILL.md (or your agent's skills folder).
name
sweep-reds
description
Repair confirmed CI failures base-first, then prepare affected PRs one at a time with exact-candidate evidence. Use when asked to sweep reds or restore the queue. Never merges, closes PRs, weakens tests, or changes CI under ordinary sweep authority.

Sweep reds: base first, then affected PRs

Read ../_shared/base-green.md and ../_shared/validation-gate.md. Success means verified required checks, not a disappearing incident or green wrapper.

Boundaries

  • Never merge, close PRs or change release-freeze state through this skill.
  • Never stash, alter the shared checkout, or touch another session's active worktree.
  • Never weaken assertions, expand tolerances, suppress failures or raise a baseline to make a check green. A proven workflow defect requires an explicitly scoped gate-repair task; ordinary sweep authorization does not authorize CI mutation.
  • Run live credentialed/paid tests, deployment or trust approvals only with the appropriate owner authorization. Missing access is HOLD, not PASS.

A. Base health

  1. Resolve live main/release refs and freeze state read-only. Confirm the intended development branch; a frozen branch remains off limits.
  2. Follow the shared exact-SHA evidence protocol. Closed/missing incidents mean UNKNOWN unless current applicable executions establish PASS. A quick observer receipt is partial; the last incident comment is only a diagnostic worklist.
  3. Reproduce each signature on an isolated pinned base. Keep setup/install failures separate from assertions. Open one owned branch/worktree per independently reviewable repair, with RED-to-GREEN proof; avoid duplicating another session's PR.
  4. Prepare/babysit the fix PR under the authorized scope, then hand it to the owner. Do not claim the base is repaired until the fix is integrated and the new exact base passes. The authorized observer, not this skill, manages its incident.
Show full SKILL.md (219 more words)Show less

B. PR queue

Inventory requested open PRs and their current head/base/candidate identities. Classify required failure, advisory failure, pending, stale, cancelled and approval needed separately. Work one PR at a time, checking ownership before each mutation.

An inherited label requires the same measured signature on the exact base, not membership in an old incident list. For stale evidence, obtain a newly constructed candidate under the current base; a rerun still uses the original run identity. Updating a contributor branch or approving a fork run needs its own authority.

An advisory check is nonblocking only if the current policy says so. Its failure may be a real product defect; advisory is not an infrastructure diagnosis. A PR with advisory failures is admissible only after all required applicable gates pass on the current candidate and the remaining risks are reported.

For proven PR defects use the /babysit repair discipline: RED, minimal fix, focused/affected suites and complete candidate validation. Preserve contributor credit. If five iterations do not converge, report the evidence and owner decision needed, then continue eligible independent items; never relabel failure as green.

Report

Record base and PR SHAs, failure signatures/counts, causal confidence, commits, run/attempt/artifact evidence, full versus partial validation and holds. Keep advisory debt visible. Hand the owner the prepared PRs and unresolved decisions; this skill never authorizes their merge.

© diegosouzapw, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in config/quality/skill-templates/sweep-reds of diegosouzapw/OmniRoute.

Open the folder on GitHubat commit f8a8131

Compare with similar skills

CI Red Sweep next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

CI Red Sweep compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
CI Red Sweep this skilldiegosouzapw/OmniRoute74k—~868Automated safety check: PassMIT
ONNX Runtime CI Managementmicrosoft/onnxruntime22k—~4.1kAutomated safety check: PassMIT
PR Babysitteropeninterpreter/openinterpreter69k3 repos~4.2kAutomated safety check: PassApache-2.0
Fix CIccusage/ccusage19k—~487Automated safety check: PassCustom licence
Azure Pipelines Log Downloaderansible/ansible71k—~825Automated safety check: PassGPL-3.0
CI Failure Triage and RepairChachamaru127/claude-code-harness3.2k1 repos~1.1kAutomated safety check: NotesMIT

Similar skills

  • ONNX Runtime CI Management

    microsoft/onnxruntime

    Official

    Triggers, re-runs and unblocks the CI checks on an ONNX Runtime pull request, after diagnosing whether a failure is transient or needs a code change.

    22k GitHub stars~4.1k tokensUpdated today
    DevOps & CloudAuto-check passed

  • PR Babysitter

    openinterpreter/openinterpreter

    Watches an open GitHub pull request until it merges, handling review comments, diagnosing CI failures and retrying flaky checks along the way.

    69k GitHub starsUsed in 3 repos~4.2k tokens
    DevelopmentAuto-check passed

  • Fix CI

    ccusage/ccusage

    Diagnoses and fixes failing GitHub Actions checks with gh. An agent skill from ccusage/ccusage.

    19k GitHub stars~487 tokensUpdated today
    DevelopmentAuto-check passed

  • Azure Pipelines Log Downloader

    ansible/ansible

    Downloads Azure Pipelines CI logs for an Ansible pull request or build so the agent can analyze test failures, after asking you first.

    71k GitHub stars~825 tokensUpdated yesterday
    DevOps & CloudAuto-check passed

  • CI Failure Triage and Repair

    Chachamaru127/claude-code-harness

    Diagnoses failing CI pipelines and tests, deciding first whether the test or the implementation is at fault, and hands hard cases to a dedicated fixer subagent.

    3.2k GitHub starsUsed in 1 repo~1.1k tokens
    DevOps & CloudAuto-check: notes

  • CI Watchdog

    latitude-dev/latitude-llm

    Continuously monitor GitHub PR CI checks and automatically fix failures until all checks pass.

    4.7k GitHub stars~1.6k tokensUpdated yesterday
    DevOps & CloudAuto-check passed

More from diegosouzapw/OmniRoute

All 50 skills in this repo

  • OmniRoute Backup and Sync CLI

    diegosouzapw/OmniRoute

    Backup and restore OmniRoute data from the CLI. Trigger incremental snapshots, sync to cloud storage, manage backup schedules, and restore from archive files.

    74k GitHub stars~948 tokensUpdated today
    Auto-check passed

  • OmniRoute Settings API

    diegosouzapw/OmniRoute

    Read and update global application settings: system prompts, thinking budget, IP filters, payload rules, combo defaults, and require-login configuration.

    74k GitHub starsUsed in 1 repo~3.6k tokens
    Auto-check passed

  • Quality Scan

    diegosouzapw/OmniRoute

    Runs a scoped, read-only quality scan on a repository candidate and reports exact evidence, failures and frozen debt, without treating a static scan as release acceptance.

    74k GitHub stars~748 tokensUpdated today
    Auto-check passed

  • OmniRoute Database Backups

    diegosouzapw/OmniRoute

    Trigger system backups, restore from backup files, and manage the SQLite database lifecycle. Supports export, import, and incremental snapshot strategies.

    74k GitHub starsUsed in 1 repo~395 tokens
    Auto-check passed

  • OmniRoute Provider Management

    diegosouzapw/OmniRoute

    Manages AI provider connections, API keys, OAuth flows and connection tests through OmniRoute's REST API across its 327-provider catalog.

    74k GitHub stars~2.4k tokensUpdated today
    Auto-check passed

  • OmniRoute Prompt Compression CLI

    diegosouzapw/OmniRoute

    Configure and test prompt compression from the CLI. Manage RTK filters, Caveman rules, stacked compression modes, and preview compression output with real…

    74k GitHub starsUsed in 1 repo~521 tokens
    Auto-check passed

Categories

Questions about CI Red Sweep

What does CI Red Sweep do?

Repairs confirmed CI failures on the base branch first, then prepares each affected PR one at a time with exact-commit evidence, never merging or weakening tests. A disciplined routine for restoring a red CI queue. Part A fixes the base: it resolves live refs read-only, reproduces each failure signature on an isolated pinned base, opens one branch or worktree per independently reviewable repair with red-to-green proof, and hands the fix PR to its owner.

When should I use CI Red Sweep?

CI Red Sweep fits situations like: restoring a red main branch before reviewing other PRs; triaging which open PRs fail for the same base reason; preparing a fix PR with red-to-green evidence; sweeping a PR queue after a CI incident.

How do I install CI Red Sweep in Claude Code?

Run `npx skills add diegosouzapw/OmniRoute --skill sweep-reds -a claude-code`. Or copy the skill folder (config/quality/skill-templates/sweep-reds in diegosouzapw/OmniRoute) into .claude/skills/sweep-reds in your project. Claude Code loads it when a task matches its description.

How do I install CI Red Sweep in Codex?

Run `npx skills add diegosouzapw/OmniRoute --skill sweep-reds -a codex`. Or copy the skill folder (config/quality/skill-templates/sweep-reds in diegosouzapw/OmniRoute) into .agents/skills/sweep-reds in your project. Codex loads it when a task matches its description.

Can I use CI Red Sweep in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add diegosouzapw/OmniRoute --skill sweep-reds -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/sweep-reds, .gemini/skills/sweep-reds, .github/skills/sweep-reds and .opencode/skills/sweep-reds in your project.

What does CI Red Sweep need to run?

SKILL.md names no scripts, command-line tools or credentials: CI Red Sweep is instructions for the agent only. Our summary lists: Read access to the repository's CI results and open PRs; The shared base-green.md and validation-gate.md files.

Does CI Red Sweep access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is CI Red Sweep safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does CI Red Sweep use?

CI Red Sweep is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does CI Red Sweep use?

About 868 tokens (SKILL.md is roughly 3.5k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to CI Red Sweep?

Skills that share tags, products or a category with CI Red Sweep: ONNX Runtime CI Management (microsoft/onnxruntime, 22k stars), PR Babysitter (openinterpreter/openinterpreter, 69k stars), Fix CI (ccusage/ccusage, 19k stars) and Azure Pipelines Log Downloader (ansible/ansible, 71k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains CI Red Sweep?

diegosouzapw (a GitHub user) maintains it in diegosouzapw/OmniRoute, which has 74,003 GitHub stars. The repository holds 50 skills in this directory. The repository was last updated on October 8, 2026.

Source: diegosouzapw/OmniRoute on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.