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-idp-initiated-flow-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-idp-initiated-flow-trust)<a href="https://agentmods.dev/skills/unboundcompute/security-agent-skills/auditing-idp-initiated-flow-trust"><img src="https://agentmods.dev/badge/skills/unboundcompute/security-agent-skills/auditing-idp-initiated-flow-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-idp-initiated-flow-trust"><img src="https://agentmods.dev/badge/skills/unboundcompute/security-agent-skills/auditing-idp-initiated-flow-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 | $0.00204 | $0.02350 |
| Opus 5.5 | $0.00082 | $0.00940 |
| Sonnet 5.5 | $0.00041 | $0.00470 |
| Haiku 4.5 | $0.00020 | $0.00235 |
Grade A, and why
auditing-idp-initiated-flow-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 1mo 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 — 146 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Auditing IdP-initiated flow trust: an unsolicited login has no request to anchor it, so validate everything else
In service-provider-initiated single sign-on the application starts the flow, so it has a request on file to bind the returning assertion to: a request id it can require the response to reference, and request state it can check. IdP-initiated single sign-on removes that anchor. Login begins at the identity provider and the application receives an assertion it never asked for, with no matching request, no InResponseTo to check, and no request state to compare against. Everything the request would have pinned down, that this response answers a login this browser started, must now be validated some other way, and often is not. An unsolicited assertion an attacker captures can be replayed, or delivered to a victim, to log a browser into an attacker-chosen account. A RelayState value trusted as a redirect target becomes an open redirect or drops the user into an attacker-controlled context after login. An assertion with no audience restriction, or one too wide, is accepted by a service provider it was never meant for. And without a replay defense, one-time-use, a tight expiry, a consumed-assertion cache, the same assertion logs in again and again. The audit treats an unsolicited assertion as untrusted, checks its signature, audience, timing, and single-use, and refuses to trust RelayState as a destination. You audit this by capturing an IdP-initiated assertion and trying to replay it, redirect with it, or use it at the wrong service provider.
When to use
- An application accepts IdP-initiated single sign-on: an unsolicited SAML assertion (or similar) with no matching service-provider request.
- The assertion may be accepted with no request binding, no or a too-wide audience, or no replay defense.
- RelayState may be trusted as a redirect target.
Scope check
Test IdP-initiated flows only against applications and identity providers you own or are authorized to assess, with test accounts. Capturing and replaying assertions logs into real sessions, so use test identities and never replay an assertion or log into an account that is not yours. If you can't name the authorization, stop.
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.
- 1mo ago First seen · 146 lines · 204 tokens per session scan A 2841c3a88bbd
auditing-idp-initiated-flow-trust is a skill published in the GitHub repository UnboundCompute/security-agent-skills (5 stars, last pushed 1mo ago), licensed MIT. It adds 204 tokens to every session and 2,350 once invoked, about $0.0008 per session on Opus 5.5. 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 2026-09-05.
Other skills, from other repositories
prodcheck-review
Review this codebase against the prodcheck pre-production checklists — security, performance, scale, integrations, post-launch readiness and people. 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…
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-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.
varlock-claude-skill
Secure environment variable management ensuring secrets are never exposed in Claude sessions, terminals, logs, or git commits.