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 UnboundCompute/security-agent-skills --skill auditing-jwt-verification-trustgit clone --depth 1 https://github.com/UnboundCompute/security-agent-skillsWrote 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/unboundcompute/security-agent-skills/auditing-jwt-verification-trust)<a href="https://agentmods.dev/skills/unboundcompute/security-agent-skills/auditing-jwt-verification-trust"><img src="https://agentmods.dev/badge/skills/unboundcompute/security-agent-skills/auditing-jwt-verification-trust/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/unboundcompute/security-agent-skills/auditing-jwt-verification-trust"><img src="https://agentmods.dev/badge/skills/unboundcompute/security-agent-skills/auditing-jwt-verification-trust.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 | $0.00218 | $0.02256 |
| Opus 5 | $0.00109 | $0.01128 |
| Sonnet 5 | $0.00044 | $0.00451 |
| Haiku 4.5 | $0.00022 | $0.00226 |
Grade A, and why
auditing-jwt-verification-trust 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 12d 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 — 139 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Auditing JWT verification trust: when a signed token is verified on the token's own terms
A JSON Web Token is only as trustworthy as the verification code, and the bug is verification that trusts
what the token itself supplies: the algorithm from the header, the key from a header-named location. If the
server accepts the token's alg, an attacker switches a signature-verified token to an HMAC one signed
with the public key, or to none; if it fetches the key from the token's kid, jku, or x5u, the
attacker points it at a key they control. You audit it by resolving whether the algorithm is pinned
server-side and where the verification key comes from, then checking the claims. This is a source-code
audit: the whole value is deciding whether a reported token flaw is actually reachable in this codebase,
so the false-positive killers carry the weight. Token-generation entropy belongs to the randomness skill;
the OAuth grant dance belongs to the OIDC skill.
When to use
- You are reviewing the code path that verifies a signed token and uses its claims for an identity decision.
- You see a decode-or-verify call, an algorithms option, a key lookup by header parameter, or a claims read.
- You want to know whether an attacker can forge a token this verification call accepts.
Scope check
Audit only systems you own or are authorized to assess, and present a crafted token only against an endpoint in scope, a forged token that verifies grants real access. Adjudicate on the verification call and its options. If you can't name the authorization, stop.
The loop
- Resolve the algorithm pinning and the key source first. Read the verification call and its options:
is the accepted algorithm pinned to a fixed value server-side, or taken from the token header, and where
does the verification key come from (a static preconfigured key or keyset, or a location named by the
token's
kid,jku, orx5u). These two facts decide whether the header-driven attacks have a sink, so settle them before flagging anything.
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.
- 12d ago First seen · 139 lines · 218 tokens per session scan A 88e20dd1797f
auditing-jwt-verification-trust is a skill published in the GitHub repository UnboundCompute/security-agent-skills (5 stars, last pushed 3d ago), licensed MIT. It adds 218 tokens to every session and 2,256 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.
Other skills, from other repositories
prodcheck-review
Review this codebase against the prodcheck pre-production checklists — security, performance, scale, integrations and post-launch readiness. Use when asked to check whether a project is ready to ship, to audit an area before launch, or to work through a specific checklist. Produces evidence with file:line citations…
integrate-arcjet-guard-genkit
Integrate Arcjet security into a Genkit JS agent using @arcjet/guard — wrap ai.defineTool, put guardMiddleware on generate({ use }) for unwrapped / MCP / filesystem tools, and read a caller-owned id from generate({ context }). Use when asked to add Arcjet to genkit, rate limit its tools, screen inbound messages, or…
integrate-arcjet-guard-langgraph
Integrate Arcjet security into a LangGraph Graph API agent using @arcjet/guard — wrap tool() / StructuredTool, wrap ToolNode for unwrapped MCP tools, and read threadid for correlation. Use when asked to add Arcjet to a LangGraph StateGraph / ToolNode agent, rate limit its tools, screen inbound messages, or block…
integrate-arcjet-guard-tanstack-ai
Integrate Arcjet security into a TanStack AI chat() app using @arcjet/guard — put guardMiddleware first on chat({ middleware }) so onBeforeToolCall gates tools, and read a caller-owned id from chat({ context }). Use when asked to add Arcjet to TanStack AI, rate limit its tools, screen inbound messages, or block prompt…
integrate-arcjet-guard-claude-managed-agents
Integrate Arcjet security into Claude Managed Agents (hosted REST+SSE, beta managed-agents-2026-04-01) using @arcjet/guard — screen user.message / initialevents before sessions.events.send, and gate custom tools on agent.customtooluse. Use when asked to add Arcjet to Claude Managed Agents, rate limit custom tools, or…
linear-claude-skill
Manage Linear issues, projects, and teams.