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

Chai Presubmit Override Decisions

by openshift-eng in openshift-eng/ai-helpers

A skill your agent uses when monitoring CHAI presubmit agent decisions in BigQuery to analyze the breakdown of override, retest, and report-only actions on PR presubmit failures and estimate CI cost…

Apache-2.0Auto-check passedDatabases

Install Chai Presubmit Override Decisions

skills CLI
$ npx skills add openshift-eng/ai-helpers --skill chai-presubmit-override-decisions -a claude-code

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

GitHub CLI
$ gh skill install openshift-eng/ai-helpers chai-presubmit-override-decisions --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/openshift-eng/ai-helpers.git skills-src && mkdir -p .claude/skills && cp -r skills-src/plugins/bigquery-ci-data/skills/chai-presubmit-override-decisions .claude/skills/chai-presubmit-override-decisions && 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
chai-presubmit-override-decisions
GitHub stars
120
Token cost
~2.5k tokens
SKILL.md length
789 words
Files
1
Skills in repo
118
Repo updated
First seen
Licence
Apache-2.0

At a glance

A skill your agent uses when monitoring CHAI presubmit agent decisions in BigQuery to analyze the breakdown of override, retest, and report-only actions on PR presubmit failures and estimate CI cost…

  • Works in 7 steps: Active repos — distinct repos with… → Per-repo decision breakdown — count and… → CI savings estimate — total overrides *… → …
  • Monitoring CHAI presubmit agent decisions in BigQuery to analyze the breakdown of override
  • SKILL.md covers When to Use This Skill, Default Timespan, CI Cost Savings and Table, plus 2 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Chai Presubmit Override Decisions is an agent skill from openshift-eng/ai-helpers. Use when monitoring CHAI presubmit agent decisions in BigQuery to analyze the breakdown of override, retest, and report-only actions on PR presubmit failures and estimate CI cost savings

Its SKILL.md is about 2.5k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Databases, covering Cloud cost optimization and Data warehousing. It works with Google BigQuery. The repository describes itself as: Developer productivity tools for Claude Code & other AI assistants. The licence is Apache-2.0.

When your agent uses it

  • Monitoring CHAI presubmit agent decisions in BigQuery to analyze the breakdown of override
  • Report-only actions on PR presubmit failures and estimate CI cost savings

Example prompts

  • “/chai-presubmit-override-decisions”

Workflow steps

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

  1. Active repos — distinct repos with decisions in the time window, listed by name. Presubmit override decisions are only enabled on select…
  2. Per-repo decision breakdown — count and percentage for each decision type per repo. Call out repos that look different from others (e.g…
  3. CI savings estimate — total overrides * $3 with a note that this estimates avoided retest costs. Break down per repo if there are enough…
  4. Consistency check — are the decision ratios similar across repos? Significant divergence could indicate different job profiles or…
  5. Top jobs — which prow jobs are driving the most decisions
  6. Notable patterns — e.g. jobs with low pass rates getting overrides, revision mismatches, tests driving most decisions
  7. Linked bugs — which Jira issues are associated with overridden failures

What it can do on your machine

Read from SKILL.md and the folder at commit a627176. 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 (its code samples are sql).

    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

Chai Presubmit Override Decisions loads about 2.5k tokens when it runs. Until then it costs about 55 tokens; SKILL.md has 789 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~55
When it runs · the whole SKILL.md, loaded when a task matches
~2.5k

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 openshift-eng/ai-helpers at commit a627176, republished under its Apache-2.0 licence (© openshift-eng). 789 words, ~2,468 tokens.

Download SKILL.mdSave it as .claude/skills/chai-presubmit-override-decisions/SKILL.md (or your agent's skills folder).
name
chai-presubmit-override-decisions
description
Use when monitoring CHAI presubmit agent decisions in BigQuery to analyze the breakdown of override, retest, and report-only actions on PR presubmit failures and estimate CI cost savings

CHAI Presubmit Override Decisions

Query openshift-gce-devel.ci_analysis_us.chai_presubmit_decision_log to monitor decisions made by the CHAI agent when evaluating PR presubmit failures. The primary goal is understanding the decision breakdown: how many failures are being overridden vs retested vs just reported on, and estimating CI cost savings from overrides.

Follow the foundations skill for cost safety, caching, and execution workflow.

When to Use This Skill

  • Monitoring the decision breakdown (override / retest / report_only / no_action_required)
  • Estimating CI cost savings from overridden presubmit failures
  • Reviewing decisions for a specific PR, repo, or job
  • Understanding why the agent made a particular decision (rationale)
  • Tracking decision trends over time
  • Auditing whether overrides correlate with post-merge regressions

Default Timespan

Use one week unless the user specifies otherwise.

CI Cost Savings

Each overridden presubmit failure saves an estimated $3 per job run in CI costs (cloud compute for the retest that was avoided). When reporting override counts, always calculate and display the estimated savings:

estimated_savings = override_count * $3.00

Table

ci_analysis_us.chai_presubmit_decision_log

Project: openshift-gce-devel

Each row represents one decision the CHAI agent made about a PR presubmit failure.

ColumnTypeNotes
logged_atTIMESTAMPWhen the record was appended (UTC)
personaSTRINGThe persona that appended the record
session_idSTRINGThe conversation session ID
repoSTRINGThe PR's repository
pr_numberINTEGERThe PR number
head_shaSTRINGThe live PR HEAD SHA the decision was made against
run_shaSTRINGThe SHA the failed run tested, when known
revision_matchBOOLEANWhether the failed run tested the live PR HEAD SHA
status_contextSTRINGFull GitHub status context (e.g. ci/prow/e2e-gcp-ovn-upgrade)
prow_job_nameSTRINGFull Prow job name
prow_build_idSTRINGProw build ID of the failed run
prow_job_urlSTRINGProw job URL of the failed run
run_finished_atTIMESTAMPWhen the failed run finished
decisionSTRINGThe action taken: override, retest, report_only, no_action_required
job_classificationSTRINGWhether the job is an override-eligible long-running e2e or integration job
tests_executedBOOLEANWhether the failed run executed tests
failing_testsSTRING (REPEATED)The failing test names
job_pass_rateFLOATFleet-wide pass rate for this job (0-100)
linked_bugsSTRING (REPEATED)Jira bugs linked to failing tests (e.g. OCPBUGS-12345)
pr_overlapSTRINGThe PR's overlap with the failing test or tested surface
rationaleSTRINGWhy this decision was made, in a short paragraph
Decision Types
DecisionMeaning
overrideAgent overrode the failure — the PR can merge without retesting. Saves CI cost.
retestAgent triggered a retest of the job.
report_onlyAgent reported on the failure but took no action (e.g. waiting for human review).
no_action_requiredAgent determined no action was needed (e.g. failure already resolved).

A healthy system should show a mix of all decision types. The breakdown reveals whether the agent is being too aggressive (too many overrides) or too conservative (mostly report_only).

Query Patterns

Show full SKILL.md (346 more words)Show less
Decision breakdown per repo with CI savings (default query)

This is the primary query to run when the user asks about CHAI decisions without a specific question. Always break down per repo — the agent is being rolled out to repos incrementally, so per-repo consistency is a key signal.

sql
SELECT
  repo,
  COUNT(*) AS total_decisions,
  COUNTIF(decision = 'override') AS overrides,
  COUNTIF(decision = 'retest') AS retests,
  COUNTIF(decision = 'report_only') AS report_only,
  COUNTIF(decision = 'no_action_required') AS no_action,
  ROUND(COUNTIF(decision = 'override') * 100.0 / COUNT(*), 1) AS override_pct,
  ROUND(COUNTIF(decision = 'retest') * 100.0 / COUNT(*), 1) AS retest_pct,
  ROUND(COUNTIF(decision = 'report_only') * 100.0 / COUNT(*), 1) AS report_only_pct,
  COUNTIF(decision = 'override') * 3 AS estimated_savings_usd
FROM `openshift-gce-devel.ci_analysis_us.chai_presubmit_decision_log`
WHERE logged_at >= TIMESTAMP_SUB(CURRENT_TIMESTAMP(), INTERVAL 7 DAY)
GROUP BY repo
ORDER BY total_decisions DESC

After running this, always report:

  • Active repos — the count of distinct repos with decisions in the time window. Presubmit override decisions are only enabled on select repositories, and repos with no PR activity in the window won't appear, so this reflects active repos, not all enabled repos.
  • Per-repo breakdown — decision counts and percentages for each repo, calling out whether the mix looks consistent across repos or if some repos skew heavily toward one decision type
  • Total override count and estimated CI savings — sum of overrides across all repos * $3
  • Overall totals — aggregate decision counts and percentages
Overall decision summary

Use alongside the per-repo query for a quick aggregate view:

sql
SELECT
  COUNT(DISTINCT repo) AS enabled_repos,
  COUNT(*) AS total_decisions,
  COUNTIF(decision = 'override') AS overrides,
  COUNTIF(decision = 'retest') AS retests,
  COUNTIF(decision = 'report_only') AS report_only,
  COUNTIF(decision = 'no_action_required') AS no_action,
  COUNTIF(decision = 'override') * 3 AS estimated_savings_usd
FROM `openshift-gce-devel.ci_analysis_us.chai_presubmit_decision_log`
WHERE logged_at >= TIMESTAMP_SUB(CURRENT_TIMESTAMP(), INTERVAL 7 DAY)
Decision breakdown by job
sql
SELECT
  prow_job_name,
  decision,
  COUNT(*) AS count,
  ROUND(AVG(job_pass_rate), 1) AS avg_pass_rate
FROM `openshift-gce-devel.ci_analysis_us.chai_presubmit_decision_log`
WHERE logged_at >= TIMESTAMP_SUB(CURRENT_TIMESTAMP(), INTERVAL 7 DAY)
GROUP BY prow_job_name, decision
ORDER BY count DESC
Daily trend (per repo)
sql
SELECT
  DATE(logged_at) AS day,
  repo,
  COUNTIF(decision = 'override') AS overrides,
  COUNTIF(decision = 'retest') AS retests,
  COUNTIF(decision = 'report_only') AS report_only,
  COUNTIF(decision = 'no_action_required') AS no_action,
  COUNT(*) AS total,
  COUNTIF(decision = 'override') * 3 AS daily_savings_usd
FROM `openshift-gce-devel.ci_analysis_us.chai_presubmit_decision_log`
WHERE logged_at >= TIMESTAMP_SUB(CURRENT_TIMESTAMP(), INTERVAL 7 DAY)
GROUP BY day, repo
ORDER BY day, repo
Override details with rationale
sql
SELECT
  logged_at,
  repo,
  pr_number,
  prow_job_name,
  job_pass_rate,
  rationale
FROM `openshift-gce-devel.ci_analysis_us.chai_presubmit_decision_log`
WHERE logged_at >= TIMESTAMP_SUB(CURRENT_TIMESTAMP(), INTERVAL 7 DAY)
  AND decision = 'override'
ORDER BY logged_at DESC
Decisions for a specific PR
sql
SELECT
  logged_at,
  prow_job_name,
  decision,
  job_pass_rate,
  revision_match,
  rationale
FROM `openshift-gce-devel.ci_analysis_us.chai_presubmit_decision_log`
WHERE logged_at >= TIMESTAMP_SUB(CURRENT_TIMESTAMP(), INTERVAL 7 DAY)
  AND repo = @repo
  AND pr_number = @pr_number
ORDER BY logged_at DESC
Overrides on stale SHAs (revision mismatch)
sql
SELECT
  logged_at,
  repo,
  pr_number,
  prow_job_name,
  revision_match,
  head_sha,
  run_sha,
  rationale
FROM `openshift-gce-devel.ci_analysis_us.chai_presubmit_decision_log`
WHERE logged_at >= TIMESTAMP_SUB(CURRENT_TIMESTAMP(), INTERVAL 7 DAY)
  AND decision = 'override'
  AND revision_match = false
ORDER BY logged_at DESC
Failing tests driving the most decisions
sql
SELECT
  test,
  COUNT(*) AS decision_count,
  COUNTIF(decision = 'override') AS override_count,
  COUNTIF(decision = 'report_only') AS report_only_count
FROM `openshift-gce-devel.ci_analysis_us.chai_presubmit_decision_log`,
  UNNEST(failing_tests) AS test
WHERE logged_at >= TIMESTAMP_SUB(CURRENT_TIMESTAMP(), INTERVAL 7 DAY)
GROUP BY test
ORDER BY decision_count DESC
LIMIT 20
Linked bugs driving overrides
sql
SELECT
  bug,
  COUNT(*) AS decision_count,
  COUNTIF(decision = 'override') AS override_count
FROM `openshift-gce-devel.ci_analysis_us.chai_presubmit_decision_log`,
  UNNEST(linked_bugs) AS bug
WHERE logged_at >= TIMESTAMP_SUB(CURRENT_TIMESTAMP(), INTERVAL 7 DAY)
GROUP BY bug
ORDER BY override_count DESC

Reporting

When presenting results, always include:

  1. Active repos — distinct repos with decisions in the time window, listed by name. Presubmit override decisions are only enabled on select repositories; repos with no PR activity in the window won't appear.
  2. Per-repo decision breakdown — count and percentage for each decision type per repo. Call out repos that look different from others (e.g. one repo with all report_only while others have a mix of overrides)
  3. CI savings estimate — total overrides * $3 with a note that this estimates avoided retest costs. Break down per repo if there are enough overrides.
  4. Consistency check — are the decision ratios similar across repos? Significant divergence could indicate different job profiles or misconfiguration.
  5. Top jobs — which prow jobs are driving the most decisions
  6. Notable patterns — e.g. jobs with low pass rates getting overrides, revision mismatches, tests driving most decisions
  7. Linked bugs — which Jira issues are associated with overridden failures

© openshift-eng, Apache-2.0. 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 plugins/bigquery-ci-data/skills/chai-presubmit-override-decisions of openshift-eng/ai-helpers.

Open the folder on GitHubat commit a627176

Compare with similar skills

Chai Presubmit Override Decisions 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.

Chai Presubmit Override Decisions compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Chai Presubmit Override Decisions this skillopenshift-eng/ai-helpers120—~2.5kAutomated safety check: PassApache-2.0
BigQuery Slot and Cost Optimizergoogle/skills21k—~2.3kAutomated safety check: PassApache-2.0
Gke Cost Analysisgoogle/skills21k—~1.5kAutomated safety check: PassApache-2.0
Deploying On GCPancoleman/ai-design-components525—~3.9kAutomated safety check: PassMIT
Neocarta Add Source Connectorneo4j-labs/neocarta147—~1.9kAutomated safety check: PassApache-2.0
Write Script Bigquerywindmill-labs/windmill18k—~2.2kAutomated safety check: PassCustom licence

Similar skills

  • BigQuery Slot and Cost Optimizer

    google/skills

    Official

    Analyzes BigQuery slot use, query costs and execution bottlenecks from INFORMATION_SCHEMA to diagnose slow queries, slot contention and unpartitioned scans.

    21k GitHub stars~2.3k tokensUpdated yesterday
    DatabasesAuto-check passed

  • Gke Cost Analysis

    google/skills

    Official

    Answer natural language questions and perform analysis on GKE cluster and workload costs using BigQuery billing exports, cost allocation data, and live cluster monitoring metrics.

    21k GitHub stars~1.5k tokensUpdated yesterday
    DevOps & CloudAuto-check passed

  • Deploying On GCP

    ancoleman/ai-design-components

    Implement applications using Google Cloud Platform (GCP) services.

    525 GitHub stars~3.9k tokensUpdated 10 mo ago
    DatabasesAuto-check passed

  • Neocarta Add Source Connector

    neo4j-labs/neocarta

    Scaffold, build, and verify a neocarta source or format connector against the connector contract.

    147 GitHub stars~1.9k tokensUpdated yesterday
    DatabasesAuto-check passed

  • Write Script Bigquery

    windmill-labs/windmill

    MUST use when writing BigQuery queries.

    18k GitHub stars~2.2k tokensUpdated yesterday
    DatabasesAuto-check passed

  • GCP Cloud Architect

    alirezarezvani/claude-skills

    Design GCP architectures for startups and enterprises. An agent skill from alirezarezvani/claude-skills.

    28k GitHub stars~3.2k tokensUpdated 1 mo ago
    DevOps & CloudAuto-check passed

More from openshift-eng/ai-helpers

All 118 skills in this repo

  • Investigate CI Reliability

    openshift-eng/ai-helpers

    Find and independently validate actionable reliability defects across OpenShift release jobs and presubmits, then export portable issue handoffs.

    120 GitHub stars~1.9k tokensUpdated 4 days ago
    Auto-check passed

  • Address Review PR

    openshift-eng/ai-helpers

    Fetch and address all PR review comments — categorize by priority, make code changes, post replies, and push.

    120 GitHub stars~2.9k tokensUpdated 4 days ago
    Auto-check passed

  • Categorize Activity Types

    openshift-eng/ai-helpers

    Categorize Jira issues into Red Hat Sankey Activity Type categories using MCP Jira tools.

    120 GitHub stars~2.4k tokensUpdated 4 days ago
    Auto-check passed

  • Has Review Work

    openshift-eng/ai-helpers

    Decide whether a GitHub PR has unanswered authorized review comments or new required CI failures worth a follow-up agent.

    120 GitHub stars~1.9k tokensUpdated 4 days ago
    Auto-check passed

  • Must Gather Analyzer

    openshift-eng/ai-helpers

    Analyze OpenShift must-gather diagnostic data including cluster operators, pods, nodes, and network components.

    120 GitHub stars~2.3k tokensUpdated 4 days ago
    Auto-check passed

  • Payload Autodl JSON

    openshift-eng/ai-helpers

    Schema for the autodl JSON data file produced by payload-analysis for database ingestion — you must use this skill whenever generating the autodl JSON file

    120 GitHub stars~2.6k tokensUpdated 4 days ago
    Auto-check passed

Works with

Categories

Questions about Chai Presubmit Override Decisions

What does Chai Presubmit Override Decisions do?

A skill your agent uses when monitoring CHAI presubmit agent decisions in BigQuery to analyze the breakdown of override, retest, and report-only actions on PR presubmit failures and estimate CI cost…. Chai Presubmit Override Decisions is an agent skill from openshift-eng/ai-helpers.

When should I use Chai Presubmit Override Decisions?

Chai Presubmit Override Decisions fits situations like: monitoring CHAI presubmit agent decisions in BigQuery to analyze the breakdown of override; report-only actions on PR presubmit failures and estimate CI cost savings.

How do I install Chai Presubmit Override Decisions in Claude Code?

Run `npx skills add openshift-eng/ai-helpers --skill chai-presubmit-override-decisions -a claude-code`. Or copy the skill folder (plugins/bigquery-ci-data/skills/chai-presubmit-override-decisions in openshift-eng/ai-helpers) into .claude/skills/chai-presubmit-override-decisions in your project. Claude Code loads it when a task matches its description.

How do I install Chai Presubmit Override Decisions in Codex?

Run `npx skills add openshift-eng/ai-helpers --skill chai-presubmit-override-decisions -a codex`. Or copy the skill folder (plugins/bigquery-ci-data/skills/chai-presubmit-override-decisions in openshift-eng/ai-helpers) into .agents/skills/chai-presubmit-override-decisions in your project. Codex loads it when a task matches its description.

Can I use Chai Presubmit Override Decisions 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 openshift-eng/ai-helpers --skill chai-presubmit-override-decisions -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/chai-presubmit-override-decisions, .gemini/skills/chai-presubmit-override-decisions, .github/skills/chai-presubmit-override-decisions and .opencode/skills/chai-presubmit-override-decisions in your project.

What does Chai Presubmit Override Decisions need to run?

SKILL.md names no scripts, command-line tools or credentials: Chai Presubmit Override Decisions is instructions for the agent only.

Does Chai Presubmit Override Decisions 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 Chai Presubmit Override Decisions 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 Chai Presubmit Override Decisions use?

Chai Presubmit Override Decisions is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Chai Presubmit Override Decisions use?

About 2.5k tokens (SKILL.md is roughly 9.9k 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 Chai Presubmit Override Decisions?

Skills that share tags, products or a category with Chai Presubmit Override Decisions: BigQuery Slot and Cost Optimizer (google/skills, 21k stars), Gke Cost Analysis (google/skills, 21k stars), Deploying On GCP (ancoleman/ai-design-components, 525 stars) and Neocarta Add Source Connector (neo4j-labs/neocarta, 147 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Chai Presubmit Override Decisions?

openshift-eng (a GitHub organization) maintains it in openshift-eng/ai-helpers, which has 120 GitHub stars. The repository holds 118 skills in this directory. The repository was last updated on October 6, 2026.

Source: openshift-eng/ai-helpers on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.