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 agentmods add skills/n-n-code/n-n-code-skills/agent-skill-generatornpx skills add n-n-code/n-n-code-skills --skill agent-skill-generatorgit clone --depth 1 https://github.com/n-n-code/n-n-code-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/n-n-code/n-n-code-skills/agent-skill-generator)<a href="https://agentmods.dev/skills/n-n-code/n-n-code-skills/agent-skill-generator"><img src="https://agentmods.dev/badge/skills/n-n-code/n-n-code-skills/agent-skill-generator.svg" alt="Measured on agentmods" 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 | $0.00069 | $0.01849 |
| Opus 5 | $0.00034 | $0.00924 |
| Sonnet 5 | $0.00014 | $0.00370 |
| Haiku 4.5 | $0.00007 | $0.00185 |
Grade A, and why
agent-skill-generator 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 4d 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 — 142 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent skill generator
Build portable skills by default, bind them to a repository or host only when the task requires it, and spend context only on behavior the target agent cannot reliably infer.
Route by mode
Choose one primary mode and honor its side-effect boundary.
| Mode | Use when | Required output | File changes |
|---|---|---|---|
design |
Turn a problem or repeated workflow into an implementation-ready skill design | Contract, portability target, package map, side-effect boundaries, and validation plan | None |
new |
Build a skill from a defined problem or workflow | Complete skill package and validation summary | Create the agreed package; update only required registration metadata and docs |
revise |
Improve an existing skill | Focused edits, validation summary, and delta | Edit only authorized scope |
audit |
Find trigger, structure, portability, or token problems | Prioritized findings with concrete targets | None unless separately requested |
validate |
Pressure-test an existing skill | Evidence record and residual risk | None unless separately requested |
optimize-trigger |
Improve activation precision or recall | Revised trigger metadata plus adjacent-case analysis | Trigger-bearing metadata only unless broader edits are requested |
If the request mixes modes, keep read-only work read-only and state which changes are authorized before editing.
Do not rename stable skill identifiers in optimize-trigger mode unless the user explicitly requests a migration.
Shared workflow
- Define the mode, target outcome, success criteria, and output location.
- Inspect the repository, existing skill package, conversation examples, and active host for facts that remove avoidable questions.
- Resolve only decisions that materially affect triggers, runtime behavior, dependencies, side effects, portability, or packaging.
- Design or assess the smallest package that makes the workflow repeatable.
- Act within the selected mode's side-effect boundary.
- Validate with the smallest evidence combination that answers the actual risk.
- Remove token waste, review the diff or findings, and report evidence plus residual uncertainty.
What ships with it
3 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.
- 4d ago First seen · 142 lines · 69 tokens per session scan A 01d3f4718b88
agent-skill-generator is a skill published in the GitHub repository n-n-code/n-n-code-skills (4 stars, last pushed 6d ago), licensed MIT. It adds 69 tokens to every session and 1,849 once invoked, about $0.0003 per session on Opus 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 2026-08-31.
Other skills, from other repositories
web-security-auditor
Aggregate SAST, SCA, DAST, secrets, API, frontend, and backend security checks into one report.
web-frontend-designer
Scaffold frontend architecture, components, routing, and UI dependencies from project config and wireframes.
web-stack-planner
Collect project brief and stack choices, then output canonical project-config.json for downstream web-builder skills.
nmap-pentest-scans
Plan and orchestrate authorized Nmap host discovery, port and service enumeration, NSE profiling, and reporting artifacts for in-scope targets.
pentest-data-exfil-tester
Test egress and DLP controls using synthetic canary data across authorized exfiltration channels.
pentest-engagement-planner
Define legal scope, rules of engagement, and authorization artifacts for downstream pentest execution.