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-oauth-token-audience-and-scope-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-oauth-token-audience-and-scope-trust)<a href="https://agentmods.dev/skills/unboundcompute/security-agent-skills/auditing-oauth-token-audience-and-scope-trust"><img src="https://agentmods.dev/badge/skills/unboundcompute/security-agent-skills/auditing-oauth-token-audience-and-scope-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-oauth-token-audience-and-scope-trust"><img src="https://agentmods.dev/badge/skills/unboundcompute/security-agent-skills/auditing-oauth-token-audience-and-scope-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 | $0.00193 | $0.02041 |
| Opus 5 | $0.00097 | $0.01020 |
| Sonnet 5 | $0.00039 | $0.00408 |
| Haiku 4.5 | $0.00019 | $0.00204 |
Grade A, and why
auditing-oauth-token-audience-and-scope-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 — 136 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Auditing OAuth token audience and scope trust: a token is only good for what it was issued for
A bearer access token is a claim with a precise domain: it was issued by a particular authorization server, for a particular audience, carrying a particular scope, and it authorizes only what those three agree on. A resource server that checks the signature but not the audience, issuer, and scope is trusting a token far beyond what it was minted for. The confusions are specific and common. A token issued for service A is presented to service B, and if B does not verify the audience it accepts A's token as its own. A scope is treated as coarser than it is, or a missing-scope check lets any valid token reach a privileged operation. A token from a different authorization server is honored by a party that only meant to trust one issuer. Each is a token used outside its domain. The audit checks that every token is validated against the audience, issuer, and scope the operation requires, not merely that it is a well-formed, signed token. You audit this by testing whether a token good for one thing is accepted for another.
When to use
- A resource server accepts OAuth or bearer access tokens to authorize calls.
- Multiple services or audiences accept tokens, and the audience or issuer may not be verified per service.
- Scopes gate operations, and the resource server may not check the specific scope each operation requires.
Scope check
Test token trust only against services you own or are authorized to assess, on non-production tokens and accounts. Presenting tokens across audiences exercises real authorization, so use non-production credentials and never use a confused token to reach real data. If you can't name the authorization, stop.
The loop
- Establish each operation's required token domain first. Name, per operation, the audience, issuer, and scope a token must have to be authorized: this service as the audience, the trusted authorization server as the issuer, the specific scope for the action. This is the false-positive killer: a resource server that verifies all three on every call authorizes exactly the tokens it should. Name the required domain, then test what the server actually accepts.
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 · 136 lines · 193 tokens per session scan A a0ec26925eb0
auditing-oauth-token-audience-and-scope-trust is a skill published in the GitHub repository UnboundCompute/security-agent-skills (5 stars, last pushed 3d ago), licensed MIT. It adds 193 tokens to every session and 2,041 once invoked, about $0.0010 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.