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 hanamizuki/solopreneur --skill using-marketergit clone --depth 1 https://github.com/hanamizuki/solopreneurWrote 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/hanamizuki/solopreneur/using-marketer)<a href="https://agentmods.dev/skills/hanamizuki/solopreneur/using-marketer"><img src="https://agentmods.dev/badge/skills/hanamizuki/solopreneur/using-marketer/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/hanamizuki/solopreneur/using-marketer"><img src="https://agentmods.dev/badge/skills/hanamizuki/solopreneur/using-marketer.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 | $0.00106 | $0.00912 |
| Opus 5.5 | $0.00042 | $0.00365 |
| Sonnet 5.5 | $0.00021 | $0.00182 |
| Haiku 4.5 | $0.00011 | $0.00091 |
Grade A, and why
using-marketer 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 20d 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.
What it actually says
Route Marketing Work
If the current agent is marketer, complete the request inline and never
delegate, even when the request explicitly asks for marketer. This guard
overrides every routing rule below and prevents recursive delegation in both
Claude Code and Codex.
Decide the route before reading or invoking any marketing domain skill. Use only the installed skills' names and descriptions for this decision.
Route the request once, in this order:
- An explicit request to stay inline: stay inline.
- An explicit request for
marketer: delegate. - One installed marketing skill fully covers every concern and deliverable: use that skill inline, even when its workflow has multiple steps or outputs.
- Otherwise: delegate to
marketer.
After choosing the route, follow exactly one of these mutually exclusive branches:
-
If the route is inline, make zero agent calls and execute the selected skill or skills inline.
-
If the route is delegation and
marketeris known unavailable, make zero agent calls and continue inline with the smallest applicable skill set. -
If the route is delegation and
marketeris available, make exactly one named-agent call:- In Codex, call
spawn_agentonce withagent_type="marketer"andfork_turns="none". Set both fields explicitly; do not omit either field or use any other value. - In Claude Code, call
Agentonce withsubagent_type="marketer".
- In Codex, call
Never spawn a generic agent, split the brief across agents, or retry with another agent type. Send only a self-contained objective, deliverables, language, constraints, and relevant evidence or paths.
In Codex, after an accepted spawn, treat that child as the selected execution
path and wait for its result before answering. The outer liveness budget is 15
wait cycles. Each cycle calls wait_agent once with timeout_ms=60000, and
every wait call counts even when a mailbox update wakes it early. The maximum
wait is therefore 15 minutes and may be shorter after progress updates. If the
wake injects the exact canonical child's completed final result, integrate it
immediately and stop polling. Otherwise, inspect the spawned canonical path
with list_agents. A polling timeout, progress-only message, or other wake
without a completed result before the cycle budget is exhausted is not a child
failure. If the canonical path is absent from list_agents, treat it as
not_found. While the child is pending_init or running and cycles remain,
wait again: do not call
send_message, followup_task, or interrupt_agent; do not spawn, retry,
begin the delegated work inline, or deliver a final answer.
If the fifteenth non-completing cycle reports the child still pending_init or
running, call interrupt_agent exactly once, mark the delegation failed, make
no further agent calls, and continue inline with the smallest applicable skill
set. Never claim that this budget fallback completed the delegation; it cannot
satisfy live delegation acceptance.
Integrate only a completed child result. If the exact named-agent call is
rejected, the canonical path is absent, or the tool reports the child as
errored or shutdown, make no further agent calls and continue inline with
the smallest applicable skill set. Explicit user cancellation or replacement
may interrupt a running child before the budget; then follow the new user intent
without continuing the original delegated work inline or claiming that
delegation completed. If the child becomes interrupted without either of
those parent-initiated reasons, make no further agent calls, surface delegation
failure, and do not continue the original work inline or claim success.
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.
- 20d ago First seen · 80 lines · 106 tokens per session scan A 007ed867b27e
using-marketer is a skill published in the GitHub repository hanamizuki/solopreneur (154 stars, last pushed 4d ago), licensed MIT. It adds 106 tokens to every session and 912 once invoked, about $0.0004 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 2026-09-27.
Other skills, from other repositories
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…