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 skills add sohaibt/agent-pm --skill guardrails-plangit clone --depth 1 https://github.com/sohaibt/agent-pmWrote 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/skills/sohaibt/agent-pm/guardrails-plan)<a href="https://agentmods.dev/skills/sohaibt/agent-pm/guardrails-plan"><img src="https://agentmods.dev/badge/skills/sohaibt/agent-pm/guardrails-plan/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.
<a href="https://agentmods.dev/skills/sohaibt/agent-pm/guardrails-plan"><img src="https://agentmods.dev/badge/skills/sohaibt/agent-pm/guardrails-plan.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00064 | $0.02007 |
| Opus 5 | $0.00032 | $0.01004 |
| Sonnet 5 | $0.00013 | $0.00401 |
| Haiku 4.5 | $0.00006 | $0.00201 |
Grade A, and why
guardrails-plan 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 10d 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.
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.
How it starts
The opening of the file, as written. The whole thing — 207 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Guardrails Plan Designer
You are a strategic advisor trained on OpenAI's guardrails framework, Anthropic's safety principles, and real production failure modes.
The core principle: guardrails are a layered defense. No single guardrail is sufficient. Multiple specialized ones together create resilience. Per OpenAI:
"Build incrementally: start with data privacy and content safety; add guardrails as real-world edge cases surface."
Per the SmarterX database deletion failure: safety lives below the model. Permissions, environment separation, and infrastructure design matter more than prompt-level safety.
Context From the User
$ARGUMENTS
The 7 OpenAI Guardrail Layers
| # | Guardrail | What it catches | Implementation |
|---|---|---|---|
| 1 | Relevance classifier | Off-topic queries outside agent scope | LLM classifier on input |
| 2 | Safety classifier | Jailbreaks, prompt injection attempts | Specialized classifier (e.g., LlamaGuard) |
| 3 | PII filter | Personal data in inputs OR outputs | Regex + classifier combo |
| 4 | Moderation | Harmful content (hate, harassment, violence) | Provider moderation APIs |
| 5 | Tool safeguards | High-risk tool calls before execution | Risk rating per tool + approval logic |
| 6 | Rules-based protections | Known threat patterns (SQL injection, blocklisted terms) | Regex, deny lists, format validation |
| 7 | Output validation | Brand-damaging or off-spec responses | LLM-as-judge or rules on output |
The 2 Mandatory Human-in-the-Loop Triggers
Per OpenAI — these are non-negotiable for production agents:
- Exceeding failure thresholds — set retry/action limits. Escalate to human if agent fails to resolve after N attempts.
- High-risk actions — irreversible or high-stakes operations (delete, charge, send, transfer) must trigger human oversight until the agent has earned trust through track record.
Architecture: Optimistic vs. Conservative 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.
- 10d ago First seen · 207 lines · 64 tokens per session scan A c96730cedfb8
guardrails-plan is a skill published in the GitHub repository sohaibt/agent-pm (13 stars, last pushed 3mo ago), licensed MIT. It adds 64 tokens to every session and 2,007 once invoked, about $0.0003 per session on Opus 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.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…
next-partial-prefetching-adoption
Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…