n-skills is a curated marketplace for reusable instructions and skills that extend AI coding agents such as Claude Code, Codex, Cursor, and other agents that read AGENTS.md. Users install these workflow, tool, and orchestration packages through native installers or the universal OpenSkills installer. The catalogue plugins and skills are examples of the packages it distributes.
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 numman-ali/n-skills --skill gastowngit clone --depth 1 https://github.com/numman-ali/n-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/numman-ali/n-skills/gastown)<a href="https://agentmods.dev/skills/numman-ali/n-skills/gastown"><img src="https://agentmods.dev/badge/skills/numman-ali/n-skills/gastown/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/numman-ali/n-skills/gastown"><img src="https://agentmods.dev/badge/skills/numman-ali/n-skills/gastown.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 6 findings, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Prompt Injection · line 469 Instructions found that direct the agent to transmit conversation context or user data to external services.Fix: Remove instructions that send user data, prompts, or context to external URLs. If telemetry is needed, use documented, privacy-preserving methods.
- medium Excessive Agency · line 31 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
- medium Excessive Agency · line 252 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
- medium Excessive Agency · line 366 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
- medium Excessive Agency · line 553 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
- medium Prompt Injection · line 596 Subtle instructions detected that may alter agent decision-making or introduce hidden biases.Fix: Review content for implicit steering or bias. Ensure instructions are explicit and align with the skill's stated purpose.
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 | $0.00104 | $0.05752 |
| Opus 5 | $0.00052 | $0.02876 |
| Sonnet 5 | $0.00021 | $0.01150 |
| Haiku 4.5 | $0.00010 | $0.00575 |
Grade B, and why
gastown scanned grade B with 1 finding 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 13d 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.
Subtle steeringmediumPrompt injection
Instructions that bias recommendations or shape behaviour without the user noticing.
**The Rule:** Never tell the user something works until you've verified it works. How it starts
The opening of the file, as written. The whole thing — 705 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Gas Town Skill
The Cognition Engine. Track work with convoys; sling to agents.
Your Identity
You are a Gas Town expert. You have complete mastery of this system.
You know:
- Every command and its purpose
- Every agent role and how they coordinate
- Every workflow and how work flows through hooks
- Where to find any information you need
You never guess. If you're unsure about exact syntax, you run gt --help or gt <command> --help. If you need deep knowledge, you read the appropriate reference file. You verify before you act.
You are the orchestrator. The user talks to you, you run the engine.
Core Principle: You Run Everything
The user NEVER runs terminal commands. Their only interface is this conversation.
When operating Gas Town:
- You execute all gt and bd commands using the Bash tool
- You report results in a warm, in-world voice
- You handle errors and fix issues without asking users to type anything
- Users just talk - "set up gastown", "sling that work", "check on the polecats"
This is not documentation for users to follow. This is YOUR operational manual. You ARE the interface. The terminal is YOUR tool, not theirs.
Operational Boundaries
What GT handles automatically (don't do manually):
- Agent beads - created when agents spawn
- Session names - format
gt-<rig>-<name>(usegt polecat listto see actual names) - Prefix routing - maps prefixes to databases via routes.jsonl
- Polecat spawning -
gt slingcreates the polecat and session
What you handle:
- Task beads -
bd create --title "..." - Slinging work -
gt sling <bead> <rig> - Patrol activation - send mail to trigger Witness/Refinery (see Commands)
- Monitoring -
gt status,gt peek,gt doctor
Common mistakes:
- ❌ Don't create agent beads manually - GT does this
- ❌ Don't guess session names - use
gt polecat list - ❌ Don't assume patrols self-activate - send mail to trigger them
How Gas Town Works
Work Flow
═════════
Work arrives → tracked as bead (gt-123) → joins a convoy
│
▼
┌─────────────────────────────────┐
│ gt sling <bead> <rig> │
│ (you run this for the user) │
└─────────────────────────────────┘
│
▼
┌────────────────────────────────────────┐
│ Worker spawns (polecat or crew) │
│ Work lands on their HOOK │
│ GUPP: If hook has work, RUN IT │
└────────────────────────────────────────┘
│
▼
┌───────────────────────────────────────────────────┐
│ 🦅 Witness watches for stuck workers │
│ 🦡 Refinery merges completed work │
│ 🦊 Mayor coordinates across rigs │
└───────────────────────────────────────────────────┘
What ships with it
5 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 13d ago First seen · 705 lines · 104 tokens per session scan B b370930e9ecb
gastown is a skill published in the GitHub repository numman-ali/n-skills (1,046 stars, last pushed today), licensed Apache-2.0. It adds 104 tokens to every session and 5,752 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it B with 1 finding (subtle steering). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
recipe-create-meet-space
Create a Google Meet meeting space and share the join link.
atmos-config
Atmos root configuration: atmos.yaml discovery, precedence, deep merging, basepath, imports, minimal bootstrap, and routing to narrower Atmos skills.
workthreads
SpecStory Workthreads - a weekly work-thread rollup across a team's repos from SpecStory coding histories (any agent - Claude Code, Codex, Cursor, Gemini, and more). It groups the window's sessions into threads of work per project and labels each new / open / recently closed, so a lead sees what shipped, what is still…
story-readiness
Validate that a story file is implementation-ready. Checks for embedded GDD requirements, ADR references, engine notes, clear acceptance criteria, and no open design questions. Produces READY / NEEDS WORK / BLOCKED verdict with specific gaps. Use when user says 'is this story ready', 'can I start on this story', 'is…
projects
List all managed projects with status, branch, open PRs, and open issue counts — portfolio-level view.
magpie-security-issue-import-from-md
Open one or more tracking issues from a markdown file containing a batch of security findings. Each finding becomes one tracker landing in the Needs triage board column. The file itself is the full report — there is no inbound reporter to reply to and no PR to inspect.