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 hunting-non-human-identity-and-secret-reachabilitygit 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/hunting-non-human-identity-and-secret-reachability)<a href="https://agentmods.dev/skills/unboundcompute/security-agent-skills/hunting-non-human-identity-and-secret-reachability"><img src="https://agentmods.dev/badge/skills/unboundcompute/security-agent-skills/hunting-non-human-identity-and-secret-reachability/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/hunting-non-human-identity-and-secret-reachability"><img src="https://agentmods.dev/badge/skills/unboundcompute/security-agent-skills/hunting-non-human-identity-and-secret-reachability.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 | $0.00151 | $0.01566 |
| Opus 5.5 | $0.00060 | $0.00626 |
| Sonnet 5.5 | $0.00030 | $0.00313 |
| Haiku 4.5 | $0.00015 | $0.00157 |
Grade A, and why
hunting-non-human-identity-and-secret-reachability 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 — 127 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Hunting non-human identity and secret reachability: existence is not the bug, reach is
A secret scanner tells you a credential is present. That is a lead, not a finding. The questions that decide severity are different: is this credential still valid, how much does it grant, and can an attacker actually reach it? A dead key in old history is noise. A live, broadly scoped token that an untrusted path leads to is an incident. Machine identities now vastly outnumber human ones, so the volume is overwhelming; the method that matters is separating reachable and powerful from merely present.
When to use
- You are reviewing secret exposure or the machine identities in a system.
- A scanner reports many secrets and you need to know which are exploitable.
- You want the blast radius of a leaked credential, not just its location.
Scope check
Inventory and validate credentials in systems you own or are authorized to test, and test validity only against your own accounts. If you can't name the authorization, stop.
The loop
-
Inventory the machine identities and secrets. Enumerate credentials across source, configuration, infrastructure definitions, build output, and runtime surfaces: keys, service-account credentials, and long-lived tokens. Record where each lives and which identity it belongs to. This is the candidate set, not the finding set.
-
Determine reachability. For each credential, ask who can reach where it sits. A secret in a public location, in client-delivered code, in a log, or on a surface an untrusted request touches is reachable. One held only in a restricted store that no untrusted path leads to is not. Reachability is the first filter that turns a lead into a candidate worth validating.
-
Establish whether it is live. A credential matters only if it still works. Confirm validity against your own account with a benign call, or from provenance (rotation records, expiry) when a live check is out of scope. A revoked or expired secret is a documented kill, however exposed it was.
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 · 127 lines · 151 tokens per session scan A dfe36fd33291
hunting-non-human-identity-and-secret-reachability is a skill published in the GitHub repository UnboundCompute/security-agent-skills (5 stars, last pushed 1mo ago), licensed MIT. It adds 151 tokens to every session and 1,566 once invoked, about $0.0006 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.