eval-fix-loop SKILL.md skillA

eval-fix-loop from gke-labs/kube-agents is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.

Install it, pinned
mdr add gke-labs/kube-agents/eval-fix-loop@git:20261105.5fd8319

Pins git:20261105.5fd8319 in mdr.lock. Pin the fingerprint instead.

Audit grade
A, 17 of 17 checks passed
Kind
SKILL.md skill
Repository
gke-labs/kube-agents
Path
.agents/skills/eval-fix-loop/SKILL.md
Versions recorded
2
Latest version
git:20261105.5fd8319, committed Nov 5, 2026
First version recorded
Nov 3, 2026
Size
14,313 bytes
Repository stars
124
License
Apache-2.0

Take one failing eval case (or one root cause behind several), diagnose it from its recorded runs, classify the miss, record a red, make the smallest allowed fix, prove three greens, and open the pull request. Use when a nightly or local run shows a red that someone must turn green, or when a repair list names a root cause.

Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.

What the file says

# Eval fix loop

One root cause per loop. A classified cause with quoted evidence before any code. Never loosen a check to fit a model's wording. Three greens on the final head. Markers flip in the same pull request.

## First-day addendum (2026-11-02)

1. Before anything, take one row from the batch file the planner names (`~/.claude/handoffs/cert-batch-<date>.md`, outside the repository) and write your session name in its claims column.
2. Read the store state first: `expected_fail` on `main`, passes and fails per model, `fail` against `not_applicable`.
3. Search open issues and open pull requests three ways: the case name, the failing check's reason string, the mechanism. A fix in flight ends the loop with a page row note.
4. The red is on the record: quote the store rows and two failing run directories, and replay them through the check as it is on `main`.
5. For a check-only fix the green is the replay of every recorded answer with the new check: all addressed fails pass, zero passes turn to fails. Three live greens are for an agent or product change only.
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint
mdr add gke-labs/kube-agents/eval-fix-loop@sha256:cdbab9933dbcdb3b

The line at the top of this page pins the label git:20261105.5fd8319, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_mcuz5ivbydlovx4p.svg)](https://markdownregistry.com/a/art_mcuz5ivbydlovx4p)

2 badge image fetches in 30 days. What the badge shows.

Versions

The registry has recorded 2 versions of this file, between Nov 3, 2026 and Nov 5, 2026, about one every day.

versioncommittedcommitsizeaudit
git:20261105.5fd8319 latest2026-11-05 5fd8319 14,313 BA view · diff
git:20261103.43f1c7c2026-11-03 43f1c7c 10,672 BA view

Audit of the latest version

A  17 of 17 checks passed. Fixed automatic checks with no AI involved, so the same checks give the same answer every run.
  • pass: Frontmatter block present
  • pass: Frontmatter declares a name
  • pass: Frontmatter declares a description
  • pass: Size between 200 bytes and 200 KB (14313 bytes)
  • pass: No zero-width or bidi control characters
  • pass: No instruction hidden inside an HTML comment
  • pass: No link to an exfiltration or paste host
  • pass: No credential-shaped string
  • pass: No instruction to send local credentials anywhere
  • pass: No text hidden with inline styles
  • pass: No prompt-injection phrasing
  • pass: No curl or wget piped into a shell
  • pass: No recursive delete of root, home or parent
  • pass: No instruction to read or print local credentials
  • pass: No base64 blob over 200 characters
  • pass: No link to a raw IP address
  • pass: No script tag

Source

GitHub

gke-labs/kube-agents · 124 stars · license Apache-2.0 · pushed 2026-11-06 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_mcuz5ivbydlovx4p
GET https://markdownregistry.com/api/v1/resolve?ref=gke-labs/kube-agents/eval-fix-loop
GET https://markdownregistry.com/api/v1/blob/cdbab9933dbcdb3ba95de6baf2f4b357e873a56e054904a7770fcaa8b86e89f0

Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.

Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.

More from gke-labs/kube-agents

edit-mirrored-skill SKILL.md skill
gke-labs/kube-agents · .agents/skills/edit-mirrored-skill/SKILL.md · Edit, sync or start mirroring a `gke-*` skill in `agents/platform/skills/` that is copied from google/skills. Use when…
git:20261016.ac3a849 · audit A · 124 stars
install-kube-agents SKILL.md skill
gke-labs/kube-agents · .agents/skills/install-kube-agents/SKILL.md · Install, set up, or deploy kube-agents (the Kubernetes Agentic Harness) and its Platform Agent onto a GKE cluster in a…
git:20261105.ce5f74d · audit C · 124 stars
review-address-feedback SKILL.md skill
gke-labs/kube-agents · .agents/skills/review-address-feedback/SKILL.md · Address review feedback on an open pull request: triage and fix findings, reply in every thread, resolve the addressed…
git:20261023.631dcf4 · audit A · 124 stars
review-adversarial SKILL.md skill
gke-labs/kube-agents · .agents/skills/review-adversarial/SKILL.md · Reviews a change adversarially — establishes what it claims to do, hunts candidates from eleven angles, then re-derives…
git:20261023.b413863 · audit A · 124 stars
review-chat-ux SKILL.md skill
gke-labs/kube-agents · .agents/skills/review-chat-ux/SKILL.md · Reviews a diff that changes what users read in Slack or Google Chat against the chat UX rules, and reports each…
git:20261105.db3b31b · audit A · 124 stars
review-docs-drift SKILL.md skill
gke-labs/kube-agents · .agents/skills/review-docs-drift/SKILL.md · Reviews a pull request for documentation drift — finds which docs the change should have updated, verifies doc claims…
git:20261030.59401a3 · audit A · 124 stars
review-preflight SKILL.md skill
gke-labs/kube-agents · .agents/skills/review-preflight/SKILL.md · Runs the pre-PR review passes — adversarial and docs-drift, required on every change, plus chat UX when chat output…
git:20261105.db3b31b · audit A · 124 stars
review-security-k8s-admission SKILL.md skill
gke-labs/kube-agents · .agents/skills/review-security-k8s-admission/SKILL.md · Reviews Kubernetes Admission Control (Webhooks, VAP/MAP) for vulnerabilities.
git:20260522.4bbd4a0 · audit A · 124 stars
review-security-k8s-agents-audit-logs SKILL.md skill
gke-labs/kube-agents · .agents/skills/review-security-k8s-agents-audit-logs/SKILL.md · Reviews Kubernetes and application audit logging for AI agents, ensuring tamper-proof observability.
git:20260521.2bcbdef · audit A · 124 stars
review-security-k8s-agents-credentials SKILL.md skill
gke-labs/kube-agents · .agents/skills/review-security-k8s-agents-credentials/SKILL.md · Reviews AI agent Kubernetes configurations for credentials exposed to agents.
git:20260521.2bcbdef · audit A · 124 stars
review-security-k8s-agents-data-exfil SKILL.md skill
gke-labs/kube-agents · .agents/skills/review-security-k8s-agents-data-exfil/SKILL.md · Reviews Kubernetes configurations to prevent data exfiltration by AI agents.
git:20260521.2bcbdef · audit A · 124 stars
review-security-k8s-agents-firewall SKILL.md skill
gke-labs/kube-agents · .agents/skills/review-security-k8s-agents-firewall/SKILL.md · Reviews Kubernetes network and firewall configurations specifically for AI agent execution sandboxes.
git:20260903.799c7e3 · audit A · 124 stars

Every file in gke-labs/kube-agents

Browse by kind, by grade A, or by owner.