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 cosmix/loom --skill loom-threat-modelgit clone --depth 1 https://github.com/cosmix/loomWrote 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/cosmix/loom/loom-threat-model)<a href="https://agentmods.dev/skills/cosmix/loom/loom-threat-model"><img src="https://agentmods.dev/badge/skills/cosmix/loom/loom-threat-model/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/cosmix/loom/loom-threat-model"><img src="https://agentmods.dev/badge/skills/cosmix/loom/loom-threat-model.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Server-Side Request Forgery · line 131 Code accesses a cloud instance metadata endpoint (e.g. 169.254.169.254). A single request can return temporary IAM credentials, making this a high-value SSRF target for credential theft.Fix: Remove access to cloud metadata endpoints unless strictly required. If metadata is needed, restrict it (e.g. IMDSv2 with hop limit) and never expose returned credentials.
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 | $0.00026 | $0.02147 |
| Opus 5 | $0.00013 | $0.01073 |
| Sonnet 5 | $0.00005 | $0.00429 |
| Haiku 4.5 | $0.00003 | $0.00215 |
Grade B, and why
loom-threat-model scanned grade B with 1 finding 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.
Cloud metadata endpointmediumServer-side request forgery
One request to 169.254.169.254 can return temporary IAM credentials.
- **SSRF** — attacker makes your server fetch an internal URL (cloud metadata `169.254.169.254`, `localhost`, internal services). **Defense is an allowlist of permitted hosts/schemes — never a blocklist.** Blocklists are Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.
How it starts
The opening of the file, as written. The whole thing — 174 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Threat Modeling
Structured identification of what can go wrong in a design, before code exists. Answers four questions (Shostack): What are we building? What can go wrong? What are we doing about it? Did we do a good job? This skill is architecture-time analysis — for finding vulns in existing code use loom-security-scan/loom-security-audit; for auth mechanism details use loom-auth.
When
New system design, architecture review, significant feature or trust-boundary change, third-party integration, or compliance evidence. Re-run when the architecture changes — a threat model is a living document, not a one-time deliverable.
Methodologies
STRIDE — the default; apply per element of the DFD
The core technique isn't "brainstorm STRIDE" — it's walking each DFD element and each data flow crossing a trust boundary, asking which STRIDE categories apply to that element.
| Threat | Violates | Typical control |
|---|---|---|
| Spoofing | Authentication | Strong authn, mTLS, signed tokens |
| Tampering | Integrity | Signatures, HMAC, input validation, WORM logs |
| Repudiation | Non-repudiation | Audit logs, signed receipts |
| Information disclosure | Confidentiality | Encryption, least-privilege, error hygiene |
| Denial of service | Availability | Rate limits, quotas, timeouts, autoscale |
| Elevation of privilege | Authorization | AuthZ checks, sandboxing, least privilege |
Element→likely-STRIDE heuristic: external entities → S, R; processes → all six; data flows → T, I, D; data stores → T, I, D (and R if logs).
DREAD — risk scoring (use with caution)
Score Damage, Reproducibility, Exploitability, Affected users, Discoverability (1–10); risk = mean. ⚠ DREAD is widely criticized as subjective and inconsistent across raters (Microsoft dropped it). Prefer a simple Likelihood × Impact matrix, or CVSS for concrete vulns, when you need defensible numbers. Whatever the scale, rank threats to drive mitigation order.
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 First seen · 174 lines · 26 tokens per session scan B 1d40ad79e33e
loom-threat-model is a skill published in the GitHub repository cosmix/loom (54 stars, last pushed today), licensed MIT. It adds 26 tokens to every session and 2,147 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it B with 1 finding (cloud metadata endpoint). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
API Discoverability for Agents
Making self-hosted services agent-discoverable — bake in a machine-readable API description (OpenAPI spec or a minimal API.md) when building, and discover-first (spec paths, repo search) before probing when integrating.
Agent Design Principles
A checklist for designing agent personas, skills, and multi-agent pipelines that stay reliable as they grow — grounded in the 12-factor-agents principles.
workers-best-practices
Cloudflare Workers best practices for production applications. Use when writing, reviewing, or configuring Workers.
find-journalists
Build, refine, dedupe, and enrich small fit-checked journalist lists for newsjack campaigns. Uses the newsjack CLI (preferred) or the medialyst MCP for news search and journalist enrichment, and falls back to a best-effort local mode with no verified contacts; the agent owns how returned data is organized.
story-origin-check
Recover the first public timestamp and canonical major coverage for a newsjacking signal, then decide whether newer coverage is the same story, a different story, or a materially new development.
relevance-coarse-filter
Cheap, high-recall first-pass filter that removes obvious junk from a detector candidate pool before expensive story-origin research and PR judgment. Decides keep, monitoronly, or reject — never ranks, writes angles, verifies dates, or decides whether to pitch.