A toolkit for evaluating, attacking, and observing watsonx Orchestrate agents, IBM’s platform for building and running AI agents, in its local Developer Edition or hosted SaaS version.

Use it to run evaluations, create benchmark workflow files, perform red-team tests, monitor agents, and interpret measures such as task success, tool-use accuracy, routing, and retrieval quality.

It provides a structured way to check whether agents work correctly, find weaknesses before release, and inspect their behavior while running.

agent-ops — Skill for Claude Code and Codex by ibm-self-serve-assets

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent

npx skills add ibm-self-serve-assets/building-blocks --skill agent-ops

Clone the repo

git clone --depth 1 https://github.com/ibm-self-serve-assets/building-blocks

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/skills/ibm-self-serve-assets/building-blocks/agent-ops/github.svg)](https://agentmods.dev/skills/ibm-self-serve-assets/building-blocks/agent-ops)

Your own site

<a href="https://agentmods.dev/skills/ibm-self-serve-assets/building-blocks/agent-ops"><img src="https://agentmods.dev/badge/skills/ibm-self-serve-assets/building-blocks/agent-ops/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/skills/ibm-self-serve-assets/building-blocks/agent-ops"><img src="https://agentmods.dev/badge/skills/ibm-self-serve-assets/building-blocks/agent-ops.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 193 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 4,615 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits

NVIDIA SkillSpector warn 7 Sept 2026

SkillSpector: 2 findings, up to high

These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →

high Privilege Escalation · line 112
Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Fix: Remove references to credential paths. Use environment variables or secrets managers. For docs, use placeholder paths (e.g., /path/to/config). Never load .env or token files in production code paths.
medium Rogue Agent · line 94
Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Fix: Remove any persistence mechanisms (cron jobs, startup scripts, state files). Skills should not maintain state across sessions without explicit user consent.

How audits are shown

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.00193	$0.04615
Opus 5.5	$0.00077	$0.01846
Sonnet 5.5	$0.00039	$0.00923
Haiku 4.5	$0.00019	$0.00462

Measured 1mo ago against content hash 8279535e7f36, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

agent-ops scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.

The scan reads SKILL.md. This mod also ships 2 executable files (examples/loan_underwriting/make_testcases.py, setup.sh), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

ibm-bob/skills/agent-ops/SKILL.md · 210 lines

How it starts

The opening of the file, as written. The whole thing — 210 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Agent Ops (watsonx Orchestrate)

This skill drives build-time evaluation, rubric scoring, red-teaming, and trace inspection for watsonx Orchestrate agents with the ADK evaluation framework (orchestrate evaluations ..., orchestrate observability traces ...). Targets: SaaS (IBM Cloud or AWS hosted) and Developer Edition (local server). On-prem Cloud Pak for Data: evaluations are documented; the rest is best effort.

Stance: ask first, emit commands second, execute rarely. Bob auto-runs only fast, read-only diagnostics. Anything that mutates state, runs an LLM, or takes minutes is emitted as a bash block for the user to run in their terminal.

Validated against: ibm-watsonx-orchestrate[agentops] 2.18.0 with evaluation framework 1.5.2, on an IBM Cloud SaaS instance, with the multi-agent loan-underwriting system in examples/loan_underwriting/. Statements marked observed come from those runs; the ADK docs win if they disagree.

First action — one-line prereq notice, then a 3-question interview

Post this once, verbatim, on the first turn:

Before we start: this skill needs the ADK with the [agentops] extra (2.18 or newer) in a Python 3.12 venv, and an activated orchestrate environment pointing at the instance where your agent is imported. assets/PREREQUISITES.md has the details. If your setup is ready, answer the three questions below.

Expert escape hatch

If the opening message already names a concrete action ("run quick-eval on benchmarks/", "why is precision 0.56", "export trace abc123", "write a rubric for refund policy"), skip the interview, say which module you are entering in one line, and run only that module's read-only diagnostics. Ask a single fused question only if routing is genuinely ambiguous.

Interview

Q1 — Target environment

Which environment is the agent imported in? (a) SaaS on IBM Cloud (api.<region>.watson-orchestrate.cloud.ibm.com) (b) SaaS on AWS (api.<region>.watson-orchestrate.ibm.com) (c) Developer Edition (local server, orchestrate server start) (d) On-premises (Cloud Pak for Data / Software Hub) Paste the instance URL if unsure; orchestrate env add infers the auth type from it.

Read the full file on GitHub · 210 lines

Files

What ships with it

53 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

1mo ago Changed 8279535e7f36
1mo ago Changed · -65 lines · +55 tokens per session 2964ae51f115
2mo ago First seen · 275 lines · 138 tokens per session scan A cfc363c4c152

Subscribe to this mod's changes

agent-ops is a skill published in the GitHub repository ibm-self-serve-assets/building-blocks (28 stars, last pushed yesterday), licensed Apache-2.0. It adds 193 tokens to every session and 4,615 once invoked, about $0.0008 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Building agents Testing

agent-ops next to 6 skills from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
agent-opsibm-self-serve-assets/building-blocks	Skill	28	yesterday	Apache-2.0	A	193	Claude Code, Codex
imagegenopenai/codex	Skill	133k	yesterday	Apache-2.0	A	113	Codex
playwright-climicrosoft/playwright	Skill	99k	yesterday	Apache-2.0	A	19	Claude Code
local-ai-agentsmicrosoft/ai-agents-for-beginners	Skill	79k	4d ago	MIT	A	200	Claude Code, Codex
systematic-debuggingobra/superpowers	Skill	313k	3d ago	MIT	A	21	Claude Code
next-cache-components-adoptionvercel/next.js	Skill	143k	yesterday	MIT	A	95	Claude Code
next-partial-prefetching-adoptionvercel/next.js	Skill	143k	yesterday	MIT	A	103	Claude Code