securing-agentic-ai-tool-invocation

securing-agentic-ai-tool-invocation is a skill for Claude Code, Codex from adriannoes/awesome-agentic-ai. It costs 34 tokens per session (2,765 once invoked), scanned A, original, MIT.

A security guide for controlling which tools an AI agent may call and what each call is allowed to do. It covers limited permissions, user-linked identities, short-lived credentials, and human approval for risky actions.

In plain words
What is it for?
Use it to design tool allowlists, restrict arguments and permissions, bind calls to users or sessions, and require approval before sensitive operations.
Why use it?
It reduces the chance that a prompt injection, unsafe tool, or overly broad permission lets an agent delete data, spend money, or access unrelated systems.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one.

Good fit Use it to design tool allowlists, restrict arguments and permissions, bind calls to users or sessions, and require approval before sensitive operations.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/adriannoes/awesome-agentic-ai/securing-agentic-ai-tool-invocation
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 adriannoes/awesome-agentic-ai --skill securing-agentic-ai-tool-invocation
Clone the repo
git clone --depth 1 https://github.com/adriannoes/awesome-agentic-ai

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.

agentmods badge for securing-agentic-ai-tool-invocation

README.md
[![agentmods](https://agentmods.dev/badge/skills/adriannoes/awesome-agentic-ai/securing-agentic-ai-tool-invocation/github.svg)](https://agentmods.dev/skills/adriannoes/awesome-agentic-ai/securing-agentic-ai-tool-invocation)
Your own site
<a href="https://agentmods.dev/skills/adriannoes/awesome-agentic-ai/securing-agentic-ai-tool-invocation"><img src="https://agentmods.dev/badge/skills/adriannoes/awesome-agentic-ai/securing-agentic-ai-tool-invocation/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.

agentmods 80×15 button for securing-agentic-ai-tool-invocation

Your own site · 80×15
<a href="https://agentmods.dev/skills/adriannoes/awesome-agentic-ai/securing-agentic-ai-tool-invocation"><img src="https://agentmods.dev/badge/skills/adriannoes/awesome-agentic-ai/securing-agentic-ai-tool-invocation.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 34 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,765 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 pass 7 Sept 2026
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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00034 $0.02765
Opus 5 $0.00017 $0.01383
Sonnet 5 $0.00007 $0.00553
Haiku 4.5 $0.00003 $0.00277

Measured 6d ago against content hash 92989526f69a, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

securing-agentic-ai-tool-invocation 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 6d ago.

The scan reads SKILL.md. This mod also ships 1 executable file (scripts/agent.py), 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.

cursor-claude-codex/skills/anthropic-cybersecurity-skills/skills/securing-agentic-ai-tool-invocation/SKILL.md · 265 lines

How it starts

The opening of the file, as written. The whole thing — 265 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Securing Agentic AI Tool Invocation

Authorized-use-only notice: This is a defensive skill. The controls below govern how an AI agent invokes tools/plugins. Deploy them on systems you own or operate. Test guardrail bypasses only against your own agent in a non-production environment.

Overview

Autonomous (agentic) AI systems decide which tool to call, with what arguments, and when, based on model reasoning over untrusted inputs. That makes the tool-invocation boundary the highest-risk control point in an agent: a single successful prompt injection or a poisoned tool can turn the agent into a confused deputy that deletes data, sends money, or pivots into connected systems. The relevant threat is MITRE ATLAS AML.T0053 (LLM Plugin Compromise) and the OWASP Agentic AI Top 10 classes for Tool Misuse, Excessive Agency, and Privilege Compromise.

The defense is layered, defense-in-depth governance of tool calls: (1) a strict allowlist of which tools the agent may call and with which argument shapes; (2) least-privilege identity binding so each tool call runs with scoped, short-lived credentials tied to the acting user/session — not a single god-mode service account; (3) policy enforcement at the call boundary (NVIDIA NeMo Guardrails dialog/flow rails and tool guardrails, or a deterministic policy wrapper); (4) human-in-the-loop (HITL) approval for high-impact actions; and (5) audit logging of every invocation for detection. This skill implements all five with verified, runnable patterns using NeMo Guardrails and a framework-agnostic Python policy wrapper.

When to Use

  • When building or hardening an agent that can call tools with real-world side effects (email, payments, file writes, infra changes, code execution).
  • When mapping OWASP Agentic AI Top 10 controls onto an existing agent framework.
  • When you need to bound the blast radius of prompt injection / tool poisoning.
  • When a compliance or governance requirement mandates approvals and audit trails for autonomous actions.
  • During an architecture review of an agent's tool layer.

Read the full file on GitHub · 265 lines

Files

What ships with it

4 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.

  1. 6d ago First seen · 265 lines · 34 tokens per session scan A 92989526f69a

Subscribe to this mod's changes

securing-agentic-ai-tool-invocation is a skill published in the GitHub repository adriannoes/awesome-agentic-ai (57 stars, last pushed 12d ago), licensed MIT. It adds 34 tokens to every session and 2,765 once invoked, about $0.0002 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 2026-09-03.

Related

Other skills, from other repositories

agent-safety

Use when bounding an LLM agent that already runs — scoping its task domain, gating tools to least privilege, defending against prompt injection in untrusted web/email/RAG text, requiring human approval on irreversible actions, capping runtime and cost, or triaging what it already did. NOT building the loop, tools, or…

ericrisco/rsc-harness · 79 tokens

importing-a-codebase

Use when the repo holds real source code but no specs: the existing-codebase branch of setting-up-a-project, normally reached via that dispatcher, directly only when the situation is unmistakable. Not for empty workspaces (starting-a-new-project) or feature work in a specced project (brainstorming).

JetBrains/thinkrail · 69 tokens

starting-a-new-project

Use when the workspace is empty — no code yet — and the user brings a raw idea: the brand-new branch of setting-up-a-project, normally reached via that dispatcher, directly only when the situation is unmistakable. Not for features in an existing project — use brainstorming instead.

JetBrains/thinkrail · 61 tokens

todos

This chat has a shared, live TODO plan — your tasks for the conversation, which the user also edits. Read this skill and reach for the todo tools whenever a request takes more than a couple of steps. It covers the plan model (group = task, items = its steps; loose items are the user's lane), how to work it: propose…

JetBrains/thinkrail · 127 tokens

writing-workflow-skills

Use when adding a new workflow skill to pi-thinkrail-workflow, changing an existing workflow skill's role, trigger, handoff, or structure, or checking a workflow skill against the workflow system's rules. Not for authoring general-purpose skills outside this package.

JetBrains/thinkrail · 60 tokens

brainstorming

Use this BEFORE any creative or feature work: building a new feature, adding functionality, changing behavior, or making a nontrivial design decision. Turns the user's request into a validated design — recorded as a spec-graph task-spec — before any implementation. Do not skip this because a change looks small.

JetBrains/thinkrail · 65 tokens