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.
npx agentmods add instructions/dmzoneill/redhat-ai-workflow/agents-mdgit clone --depth 1 https://github.com/dmzoneill/redhat-ai-workflowWrote 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.
[](https://agentmods.dev/instructions/dmzoneill/redhat-ai-workflow/agents-md)<a href="https://agentmods.dev/instructions/dmzoneill/redhat-ai-workflow/agents-md"><img src="https://agentmods.dev/badge/instructions/dmzoneill/redhat-ai-workflow/agents-md.svg" alt="Measured on agentmods" height="20"></a>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 | $0.12050 | $0.12050 |
| Opus 5 | $0.06025 | $0.06025 |
| Sonnet 5 | $0.02410 | $0.02410 |
| Haiku 4.5 | $0.01205 | $0.01205 |
Grade C, and why
redhat-ai-workflow AGENTS.md scanned grade C with 2 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 4d ago.
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.
Reaches for credential fileshighPrivilege escalation
SSH keys, cloud credentials, git-credentials, .npmrc, /etc/shadow: reading these is how a config file becomes a credential leak.
| `~/.kube/config.s` | Stage | Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
| `curl` | `aa_curl` | `curl_get()`, `curl_post()`, `curl_put()`, `curl_delete()`, `curl_patch()` | `developer` | How it starts
The opening of the file, as written. The whole thing — 1,315 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AI Workflow Assistant
Your Role
You are an AI assistant managing software development workflows across multiple projects. Your job is to help developers with:
- Daily work: Starting issues, creating branches, making commits, opening MRs
- DevOps: Deploying to ephemeral environments, monitoring, debugging
- Incidents: Investigating alerts, checking logs, coordinating response
- Releases: Building images, promoting to environments, tracking deployments
How This System Works
config.jsondefines the projects you manage (repos, namespaces, URLs, credentials)- Personas load tool sets optimized for different work types
- Skills are pre-built workflows that chain tools together with logic
- MCP Tools are individual operations (git, jira, gitlab, k8s, etc.)
- Memory persists context across sessions (active issues, learned patterns)
Key Principles
- Use skills for common workflows (they chain tools automatically)
- Use MCP tools instead of CLI commands (they handle auth/errors)
- CLI only for running app code (
pytest,python app.py) or when no tool exists - Never hardcode project-specific values - they come from
config.json
CRITICAL: Skill-First Behavior
The Golden Rule
BEFORE attempting ANY task, ALWAYS check for a matching skill first.
Skills are tested, reliable workflows that handle edge cases. Manual steps are error-prone.
Decision Tree
When user requests an action:
1. Parse intent → What are they trying to do?
2. Check skills → Does skill_list() have a matching skill?
├─ YES → Run the skill via CallMcpTool (see syntax below)
└─ NO → Continue to step 3
3. Check persona → Do I have the right tools loaded?
├─ NO → Load persona via CallMcpTool
└─ YES → Proceed with manual steps
4. Execute → Only now attempt manual execution
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.
- 4d ago First seen · 1,315 lines · 12,050 tokens per session scan C 7822d6fb56ee
redhat-ai-workflow AGENTS.md is an instructions file published in the GitHub repository dmzoneill/redhat-ai-workflow (5 stars, last pushed today), licensed Apache-2.0. It adds 12,050 tokens to every session, about $0.0602 per session on Opus 5. A static security scan graded it C with 2 findings (reaches for credential files, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
apflow copilot-instructions.md
Copilot instructions for aiperceivable/apflow, covering project overview, core principles, python code quality, readability and types (mandatory).
swiss-snb-mcp CLAUDE.md
Claude Code instructions for malkreide/swiss-snb-mcp, covering claude.md, teil 1 — konventionen (portfolio-weit), vor der arbeit, tests and wenn etwas rot ist.
ai-cli-utils GEMINI.md
Gemini CLI instructions for sergeiwallace/ai-cli-utils, covering ai-cli-utils — gemini context, tech stack, development conventions, public open-source package standards and ai sync scope boundary.
workflows-mcp-server AGENTS.md
Instructions for cyanheads/workflows-mcp-server, covering agent protocol, what's next?, core rules, patterns and tool.
unraid GEMINI.md
Gemini CLI instructions for dinglebear-ai/unraid, a project described as: Query, monitor, and manage Unraid servers via GraphQL API through MCP tools. Supports system info, Docker, VMs, array/parity, notifications, plugins, rclone, and live telemetry.
Python standards
Apply Python quality and style conventions to all Python files.