foundry-agt

foundry-agt is a skill for Claude Code, Codex from aiappsgbb/awesome-gbb. It costs 222 tokens per session (6,620 once invoked), scanned A, original, MIT.

Middleware for Microsoft Agent Framework agents that checks messages and tool calls against written policies before actions run. Middleware is software placed between an agent and its tools to enforce rules.

In plain words
What is it for?
Use it to block disallowed actions, reject messages containing personally identifying information, record audit events, and load governance rules from YAML files.
Why use it?
It moves safety checks out of the model's instructions and into code that can allow or deny each tool call deterministically.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: mentions AGENTS.md.

Good fit Use it to block disallowed actions, reject messages containing personally identifying information, record audit events, and load governance rules from YAML files.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/aiappsgbb/awesome-gbb/foundry-agt
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 aiappsgbb/awesome-gbb --skill foundry-agt
Clone the repo
git clone --depth 1 https://github.com/aiappsgbb/awesome-gbb

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 foundry-agt

README.md
[![agentmods](https://agentmods.dev/badge/skills/aiappsgbb/awesome-gbb/foundry-agt/github.svg)](https://agentmods.dev/skills/aiappsgbb/awesome-gbb/foundry-agt)
Your own site
<a href="https://agentmods.dev/skills/aiappsgbb/awesome-gbb/foundry-agt"><img src="https://agentmods.dev/badge/skills/aiappsgbb/awesome-gbb/foundry-agt/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 foundry-agt

Your own site · 80×15
<a href="https://agentmods.dev/skills/aiappsgbb/awesome-gbb/foundry-agt"><img src="https://agentmods.dev/badge/skills/aiappsgbb/awesome-gbb/foundry-agt.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 222 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 6,620 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 $0.00222 $0.06620
Opus 5 $0.00111 $0.03310
Sonnet 5 $0.00044 $0.01324
Haiku 4.5 $0.00022 $0.00662

Measured yesterday against content hash 4f6ffb62c986, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

foundry-agt 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 yesterday.

The scan reads SKILL.md. This mod also ships 5 executable files (references/maf-middleware-snippet.py, references/python/__init__.py, references/python/capability_detector.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.

skills/foundry-agt/SKILL.md · 524 lines

How it starts

The opening of the file, as written. The whole thing — 524 lines — stays where its author put it; the contents beside it link to each section on GitHub.

foundry-agt — Microsoft Agent Governance Toolkit, in-process (Path A)

Status: AGT 4.1.0 (Public Preview, MIT) + selective Microsoft Agent Framework (MAF) packages. Exact pinned versions, the released-source SHA, and the executable validation script are the single source of truth in references/upstream-pin.md — re-run its validation.script before trusting any re-pin (AGENTS.md § 9).

AGT 4.1 is Public Preview. agt verify is toolkit self-assessment, not certification, an independent audit, or a guarantee that a workload meets every organizational or regulatory control.


Why action governance matters

A system-prompt instruction like "never call delete_account" is a request to the model, not an enforcement point — a jailbroken or mis-routed turn can still emit the tool call. AGT moves that check out of the model and into middleware that runs before the tool function executes: CapabilityGuardMiddleware inspects the real FunctionInvocationContext for every tool invocation and decides allow/deny deterministically, in-process, before call_next() (the actual tool body) ever runs. A denied call raises MiddlewareTermination and the tool body never executes at all — this is proven locally by references/python/contract_probe.py, not just asserted from upstream documentation.


Scope & ownership

For the complete per-agent adoption and release-evidence workflow, see foundry-agentops. This skill remains authoritative for AGT policy authoring and in-process runtime governance; AgentOps aggregates evidence and never replaces those enforcement controls.

Concern Owner Notes
Action / tool-call plane — may the agent invoke this named tool at all? This skill (AGT) Deterministic, in-process, pre-execution by tool name. Argument validation remains the caller's / tool-body's responsibility.
Message / token content — is this text safe to send or show? Azure AI Content Safety A different plane entirely; pair with AGT, don't substitute for it.
Edge auth, rate limiting, product policy APIM / gateway — see citadel-spoke-onboarding Gates the HTTP edge; AGT gates inside the tool loop.
Network isolation (VNet, private endpoints) foundry-vnet-deploy Network plane, not policy plane.
Quality / task-adherence evaluation foundry-evals AGT governs the safety of actions, not answer quality.
Telemetry plumbing (App Insights, OTel exporters) foundry-observability AGT emits CloudEvents; this skill owns the pipe they flow through.

Read the full file on GitHub · 524 lines

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. yesterday Changed · +5 lines 4f6ffb62c986
  2. 9d ago First seen · 519 lines · 222 tokens per session scan A 5f5e55cd22c9

Subscribe to this mod's changes

foundry-agt is a skill published in the GitHub repository aiappsgbb/awesome-gbb (5 stars, last pushed 2d ago), licensed MIT. It adds 222 tokens to every session and 6,620 once invoked, about $0.0011 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 2026-08-31.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

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…

microsoft/ai-agents-for-beginners · 200 tokens

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…

vercel/next.js · 95 tokens

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…

vercel/next.js · 83 tokens

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…

vercel/next.js · 170 tokens

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…

vercel/next.js · 103 tokens