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-path-traversal-and-file-accessgit 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-path-traversal-and-file-access)<a href="https://agentmods.dev/skills/unboundcompute/security-agent-skills/hunting-path-traversal-and-file-access"><img src="https://agentmods.dev/badge/skills/unboundcompute/security-agent-skills/hunting-path-traversal-and-file-access/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-path-traversal-and-file-access"><img src="https://agentmods.dev/badge/skills/unboundcompute/security-agent-skills/hunting-path-traversal-and-file-access.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 | $0.00163 | $0.02021 |
| Opus 5.5 | $0.00065 | $0.00808 |
| Sonnet 5.5 | $0.00033 | $0.00404 |
| Haiku 4.5 | $0.00016 | $0.00202 |
Grade A, and why
hunting-path-traversal-and-file-access 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 — 135 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Hunting path traversal and file access: when a name escapes its directory
Applications constantly turn a name into a file: a download parameter into a file to send, an upload name into a destination to write, a template name into an include, a route into a static asset. The danger is that a filesystem path is hierarchical, and a name carrying dot-dot segments, an absolute prefix, or a drive letter can climb out of the directory the code meant to stay in. When that happens a read reaches secrets, configuration, or source outside the intended base, and a write lands a file, sometimes an executable one, where it should never go. The reliable defense is to resolve the path to a canonical absolute form and then confirm it is still under the intended base, but many applications check the raw input before decoding or normalizing, which an encoded or layered payload walks straight past. You find these by locating every path built from input and asking whether confinement is verified after canonicalization.
When to use
- Untrusted input names a file to read, write, include, or send: downloads, uploads, templates, static routes.
- Archive extraction or a bulk import writes members to paths derived from the archive.
- A path is validated against a base directory and you need to confirm the check survives normalization.
Scope check
Test path traversal only against applications you own or are authorized to assess, on non-production data, because a confirming read exposes real files and a confirming write can corrupt or plant content. Read a benign marker file and write to a scratch location for any proof, and prefer an isolated instance. If you can't name the authorization, stop.
The loop
- Establish whether confinement is checked after canonicalization first. For each path built from input, find where and how the code confirms the target stays under the intended base. This is the false-positive killer: resolving to a canonical absolute path and verifying it is a descendant of the base directory holds, while a check on the raw input before decoding or normalization, or a blocklist of dot-dot, is bypassable. Name the check and when it runs relative to normalization.
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 · 135 lines · 163 tokens per session scan A 4d4ab20ea250
hunting-path-traversal-and-file-access is a skill published in the GitHub repository UnboundCompute/security-agent-skills (5 stars, last pushed 1mo ago), licensed MIT. It adds 163 tokens to every session and 2,021 once invoked, about $0.0007 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.