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 m3taz-ahmed/ai-globals --skill supply-chain-lordgit clone --depth 1 https://github.com/m3taz-ahmed/ai-globalsWrote 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/m3taz-ahmed/ai-globals/supply-chain-lord)<a href="https://agentmods.dev/skills/m3taz-ahmed/ai-globals/supply-chain-lord"><img src="https://agentmods.dev/badge/skills/m3taz-ahmed/ai-globals/supply-chain-lord/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/m3taz-ahmed/ai-globals/supply-chain-lord"><img src="https://agentmods.dev/badge/skills/m3taz-ahmed/ai-globals/supply-chain-lord.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 | $0.00033 | $0.01468 |
| Opus 5 | $0.00016 | $0.00734 |
| Sonnet 5 | $0.00007 | $0.00294 |
| Haiku 4.5 | $0.00003 | $0.00147 |
Grade A, and why
supply-chain-lord 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.
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 — 76 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Supply Chain Lord
[OBJ] Supply-chain security for AI-generated code. AI agents add imports faster than humans can review them.
Problem
AI agents commonly add imports of external packages not declared in the project manifest. This is a supply-chain attack vector: a hallucinated package name may match a malicious typosquatted package. SAST/DAST catch this downstream — by then the code is committed.
Rules
- [REQ] Dependency guard at diff time. Every diff with new imports MUST pass through
runtime/supply_chain_guard.py. Undeclared import = WARN (strict mode = BLOCK). - [REQ] Lockfile as source of truth. Declared dependencies come from
pyproject.toml/requirements.txt(Python),package.json/package-lock.json(Node),composer.json/composer.lock(PHP),go.mod(Go). Never trustimportstatements as the dependency list. - [REQ] Typosquat detection. New import names within edit distance ≤ 2 of a declared dependency = WARN (e.g.,
reqeustsvsrequests). Use Levenshtein distance. - [REQ] Pin SHAs. Container images and GitHub Actions pinned by SHA, not tag. Tags are mutable; SHAs are immutable.
uses: actions/checkout@v4= reject;uses: actions/checkout@<sha>= accept. - [REQ] SBOM generation. Every release artifact ships with an SBOM (CycloneDX or SPDX).
aizee sbom generatebefore release. No SBOM = no release. - [REQ] Cosign signing. Container images signed with Cosign (keyless OIDC). Unsigned image = BLOCK deploy. Verify signature on pull.
- [REQ] Minimum release age. New dependency versions must be published ≥ 7 days before adoption. Brand-new releases have not been vetted; a non-trivial fraction of supply-chain attacks are caught and yanked within the first few days.
- [REQ] No floating ranges. Prohibit
latest,*, unbounded>=in manifests. These auto-resolve to brand-new releases. Pin to exact versions or bounded ranges. - [REQ] Audit dependencies.
pip-audit/npm audit/composer auditbefore every release. Critical CVE = BLOCK. - [REQ] No downgrade for type errors. Never downgrade a dependency to fix a type error. Fix the code or upgrade. Downgrading = security regression.
- [REQ] Provenance verification. Verify dependency provenance (e.g.,
pip --require-hashes,npm ciwith lockfile,composer installwith lock). Nopip installwithout hash checking in CI. - [REQ] MCP server provenance. MCP servers registered as securables (
runtime/mcp_securable.py) with verified endpoints. No ungoverned MCP server in the supply chain. - [REQ] Diff scope. A diff that adds a new dependency MUST also update the lockfile in the same diff. Dependency without lockfile update = BLOCK.
- [REQ] License compliance. Every declared dependency has a license in the allowlist (MIT, Apache-2.0, BSD, ISC). GPL/AGPL in a commercial project = BLOCK (require LEGAL persona review).
- [PROHIBIT]
git add ./git add -A— stage only files you modified. - [PROHIBIT] Installing a dependency published < 7 days ago.
- [PROHIBIT] Floating version ranges (
latest,*, unbounded>=). - [PROHIBIT] Unsigned container images in production.
- [PROHIBIT] A new import without a corresponding lockfile/manifest update in the same diff.
- [REQ] MCP tool poisoning defense. Hash-pin or sign allowed MCP tool definitions. Re-validate on every list refresh. CVE-2025-54136 (CVSS 8.8) showed tool definitions can change server-side after initial review.
- [REQ] Model weight provenance. Verify model weights have provenance metadata. Unsafe serialization (pickle) must be rejected. Use safe formats (safetensors, GGUF). AI/ML SBOMs required for all model deployments.
- [REQ] MCP CVE tracking. Monitor and patch: CVE-2026-52869 (Python SDK session-ID injection), CVE-2026-52870 (cross-client task access), CVE-2026-59950 (WebSocket origin bypass), CVE-2026-25536 (TypeScript SDK data leak), CVE-2026-27896 (Go SDK key parsing). Pin SDK versions above fix releases.
- [REQ] Deadbugz campaign awareness. Malicious MCP servers may activate payloads after a call-count threshold, evading initial review. Monitor for behavioral changes in MCP servers over time.
- [REQ] Confused-deputy prevention. MCP proxies with static client_id and DCR can let malicious clients hijack consent cookies. Use CIMD (Client ID Metadata Documents) instead of DCR. Validate issuer per RFC 9207.
- [REQ] AI/ML SBOM. Every AI/ML release artifact ships with an AI-specific SBOM (model weights, training data provenance, serialization format, dependencies). CycloneDX AI extension or SPDX 3.0 AI profile.
- [PROHIBIT] Using MCP servers without hash-pinned tool definitions and CVE compliance verification.
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.
- 6d ago Changed · +7 lines 4813ba381c90
- 9d ago First seen · 69 lines · 33 tokens per session scan A 643465a7a71b
supply-chain-lord is a skill published in the GitHub repository m3taz-ahmed/ai-globals (5 stars, last pushed yesterday), licensed MIT. It adds 33 tokens to every session and 1,468 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.
Other skills, from other repositories
gke-compute-classes
Configures, optimizes, and troubleshoots GKE ComputeClasses. Use when configuring Spot VMs with on-demand fallback, targeting specific accelerators (GPUs/TPUs) or machine families, restricting ComputeClass access, or debugging pending pods related to node pool auto-creation. Do not use for cluster-level Node Auto…
gke-reliability
Improves GKE workload reliability, using PDBs, health probes, and topology spread constraints. Use when configuring GKE workload reliability, setting up PDBs, or configuring GKE health probes (liveness, readiness, startup). Don't use for disaster recovery setup or full cluster backups (use gke-backup-dr instead).
gke-workload-security
Audits, configures, and hardens workload-level security controls for Google Kubernetes Engine (GKE) applications and namespaces. Covers running cluster security audits (auditcluster.sh), configuring Workload Identity Federation (impersonation, KSA/GSA binding, and pod setup), enforcing Network Policies (default-deny…
nemo-automodel-launcher-config
Configure NeMo AutoModel job launches for interactive runs, Slurm clusters, and SkyPilot cloud execution.
azure-mgmt-botservice-dotnet
Azure Resource Manager SDK for Bot Service in .NET. Management plane operations for creating and managing Azure Bot resources, channels (Teams, DirectLine, Slack), and connection settings. Triggers: "Bot Service", "BotResource", "Azure Bot", "DirectLine channel", "Teams channel", "bot management .NET", "create bot".
cloud-architect
Designs cloud architectures, creates migration plans, generates cost optimization recommendations, and produces disaster recovery strategies across AWS, Azure, and GCP. Use when designing cloud architectures, planning migrations, or optimizing multi-cloud deployments. Invoke for Well-Architected Framework, cost…