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-scheduled-job-and-search-path-hijacksgit 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-scheduled-job-and-search-path-hijacks)<a href="https://agentmods.dev/skills/unboundcompute/security-agent-skills/hunting-scheduled-job-and-search-path-hijacks"><img src="https://agentmods.dev/badge/skills/unboundcompute/security-agent-skills/hunting-scheduled-job-and-search-path-hijacks/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-scheduled-job-and-search-path-hijacks"><img src="https://agentmods.dev/badge/skills/unboundcompute/security-agent-skills/hunting-scheduled-job-and-search-path-hijacks.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 | $0.00161 | $0.02052 |
| Opus 5.5 | $0.00064 | $0.00821 |
| Sonnet 5.5 | $0.00032 | $0.00410 |
| Haiku 4.5 | $0.00016 | $0.00205 |
Grade A, and why
hunting-scheduled-job-and-search-path-hijacks 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 — 143 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Hunting scheduled-job and search-path hijacks: what runs, as whom, over what you can write
Automation escalates privilege quietly. A job runs on a timer as a privileged identity, and if anything it touches is writable by a lesser user, or if it resolves a command through a path or a wildcard that a lesser user can influence, that user's content runs as the job's identity. The bug is never the schedule; it is the trust the job places in a writable script, a writable directory on its search path, an unqualified command name, or a filename that a wildcard hands to a command as an option. You find it by listing every job, the identity it runs as, and every input it trusts, then checking which of those inputs a lesser user controls.
When to use
- You are auditing a host or image for local privilege escalation through automation.
- Periodic jobs, timers, or service-triggered scripts run as a privileged identity.
- Those jobs execute scripts, resolve commands by name, or expand wildcards over directories.
Scope check
Audit scheduled-job and path escalation only on hosts or images you own or are authorized to test, with an unprivileged account you may escalate from. If you can't name the authorization, stop.
The loop
-
Enumerate every job and the identity it runs as. Inventory periodic jobs, timers, and service-triggered scripts across every location they can be defined, system-wide and per-user, and record which identity each runs as. A job running as an unprivileged service account is a smaller prize than one running as root, but map them all; the identity sets the stakes.
-
List every input each job trusts. For each job, follow what it executes and reads: the job script itself, any script or config or data file it sources or parses, the directories it operates over, and the commands it launches. Each of these is an input the job trusts to be benign. The audit is which of those inputs a lesser user can change.
-
Check writability of scripts, configs, and directories. For every trusted input, check whether a lower-privileged user can write it: a job script that is group- or world-writable, a config sourced from a writable path, a directory the job writes into or globs over that a lesser user owns. A writable trusted input is a direct hijack; the lesser user plants content that runs as the job's identity.
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 · 143 lines · 161 tokens per session scan A 7d8df52072f7
hunting-scheduled-job-and-search-path-hijacks is a skill published in the GitHub repository UnboundCompute/security-agent-skills (5 stars, last pushed 1mo ago), licensed MIT. It adds 161 tokens to every session and 2,052 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.