talk-tal-skills-security

talk-tal-skills-security is a skill for Codex from jscraik/Agent-Skills. It costs 45 tokens per session (1,718 once invoked), scanned A, original, Apache-2.0.

A defensive review guide for checking AI-agent skills, plugins, and tools before adoption. It examines where they came from, what they can access, and how they handle data and approvals.

In plain words
What is it for?
Reviewing third-party agent add-ons, creating intake checklists, and deciding what must be validated before use.
Why use it?
It helps reveal unsafe permissions, unclear origins, data exposure, and weak sandboxing before a tool is trusted.

Skill for Codex

Written for Codex: agents/openai.yaml present.

Good fit Reviewing third-party agent add-ons, creating intake checklists, and deciding what must be validated before use.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/jscraik/agent-skills/talk-tal-skills-security
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 jscraik/Agent-Skills --skill talk-tal-skills-security
Clone the repo
git clone --depth 1 https://github.com/jscraik/Agent-Skills

Made for: 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 talk-tal-skills-security

README.md
[![agentmods](https://agentmods.dev/badge/skills/jscraik/agent-skills/talk-tal-skills-security/github.svg)](https://agentmods.dev/skills/jscraik/agent-skills/talk-tal-skills-security)
Your own site
<a href="https://agentmods.dev/skills/jscraik/agent-skills/talk-tal-skills-security"><img src="https://agentmods.dev/badge/skills/jscraik/agent-skills/talk-tal-skills-security/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 talk-tal-skills-security

Your own site · 80×15
<a href="https://agentmods.dev/skills/jscraik/agent-skills/talk-tal-skills-security"><img src="https://agentmods.dev/badge/skills/jscraik/agent-skills/talk-tal-skills-security.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 45 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,718 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.
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 $0.00045 $0.01718
Opus 5 $0.00023 $0.00859
Sonnet 5 $0.00009 $0.00344
Haiku 4.5 $0.00005 $0.00172

Measured 8d ago against content hash 61908b745d5e, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

talk-tal-skills-security 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 8d 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.

Plugins/aidevcon/skills/talk-tal-skills-security/SKILL.md · 129 lines

How it starts

The opening of the file, as written. The whole thing — 129 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Skills Security — Liran Tal

Liran Tal explains why AI-agent skills need dependency-style review. Use this skill to summarize the talk, build defensive review checklists, assess skill-governance gaps, and design safer intake processes for third-party skills, tools, and plugins.

When To Use

Use this skill for a transcript-grounded explanation of Tal's defensive model or for a bounded review of an existing skill, plugin, tool, or adoption process. Do not use it to reconstruct offensive live-demo mechanics or to treat the talk as proof of a repository's current behavior.

Inputs

  • User question or the named review target.
  • outline.md, quote.md, and the relevant passage in transcript.md.
  • For an application review, the authorized local truth surface and any existing validation output.

Outputs

  • A grounded talk explanation with clear attribution and a defensive boundary.
  • For a review, an evidence-qualified checklist, adoption decision, and the smallest next remediation or proof step.

Workflow

  1. Read outline.md to establish scope.
  2. Use quote.md only to find a candidate claim or concise advisory excerpt.
  3. Verify every factual, detailed, or quoted claim against the relevant transcript.md passage before attributing it to Tal. If the published transcript omits the detail, say so and use only the closest safe principle.
  4. State the review target before assessing it: canonical source, packaged artifact, runtime projection, or proposed permission/action. Record the source, version or ref, and digest when available. Evidence for one target does not prove another.
  5. Inspect the named target and its direct evidence surfaces first. Widen the search only when those surfaces cannot answer the review question.
  6. Keep talk evidence and local-project evidence visibly separate. Mark each local finding present, gap, or unknown; do not infer repository behavior from the talk.
  7. Stop at the first required review lane that fails or is blocked. Give the smallest remediation or proving step for that lane before widening the review.

Read the full file on GitHub · 129 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. 8d ago First seen · 129 lines · 45 tokens per session scan A 61908b745d5e

Subscribe to this mod's changes

talk-tal-skills-security is a skill published in the GitHub repository jscraik/Agent-Skills (8 stars, last pushed 10d ago), licensed Apache-2.0. It adds 45 tokens to every session and 1,718 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.