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-jit-provisioning-and-role-mappinggit 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-jit-provisioning-and-role-mapping)<a href="https://agentmods.dev/skills/unboundcompute/security-agent-skills/auditing-jit-provisioning-and-role-mapping"><img src="https://agentmods.dev/badge/skills/unboundcompute/security-agent-skills/auditing-jit-provisioning-and-role-mapping/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-jit-provisioning-and-role-mapping"><img src="https://agentmods.dev/badge/skills/unboundcompute/security-agent-skills/auditing-jit-provisioning-and-role-mapping.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 | $0.00197 | $0.02452 |
| Opus 5.5 | $0.00079 | $0.00981 |
| Sonnet 5.5 | $0.00039 | $0.00490 |
| Haiku 4.5 | $0.00020 | $0.00245 |
Grade A, and why
auditing-jit-provisioning-and-role-mapping 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 — 152 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Auditing JIT provisioning and role mapping: the first login decides the role, from claims you did not issue
Just-in-time provisioning creates a local account the first time someone logs in through a federated identity provider, and often updates that account's roles on every subsequent login, all from the claims in the login assertion. The claims, group memberships, an email, a domain, a department, come from the identity provider, not from you, so JIT hands account creation and role assignment to whatever the assertion says. That is fine when the mapping is tight and the claims are validated, and dangerous when either is loose. A claim-to-role mapping that grants more privilege than the claim should, or defaults every new user into a privileged role, elevates at first login. An email or domain claim trusted to auto-join a tenant or organization lets an attacker with a lookalike or attacker-controlled address land inside it. A JIT update that re-reads mutable claims every login can re-elevate or reassign an account an admin previously corrected. And any claim provisioned without validation, an unverified email, a group the attacker can influence, an unsigned or overtrusted assertion field, becomes an access grant. The audit follows a federated login into the account and roles it creates and checks that every claim is validated and every mapping is minimal. You audit this by logging in with shaped assertions and seeing what account and privilege you are given.
When to use
- A federated login (SAML or OIDC) provisions a local account and assigns roles or tenancy from assertion claims on first login, and may update them on repeat logins.
- A claim-to-role mapping may over-grant, default new users into a privileged role, or trust an email or domain claim to auto-join an organization.
- Claims (email, groups, department) may be provisioned without validation.
Scope check
Test JIT provisioning only against applications and identity providers you own or are authorized to assess, on non-production tenants. Logging in with shaped assertions creates real accounts and role grants, so use test identity providers and tenants and never provision or elevate access in an organization 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 · 152 lines · 197 tokens per session scan A f5122f2feaf2
auditing-jit-provisioning-and-role-mapping is a skill published in the GitHub repository UnboundCompute/security-agent-skills (5 stars, last pushed 1mo ago), licensed MIT. It adds 197 tokens to every session and 2,452 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.