메뉴
The foundational source of truth for what systemprompt.io is, who it serves, and how it goes to market. Load this skill before any other content skill. Defines product identity, ICP, go-to-market strategy, competitive positioning, and messaging hierarchy.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Maintains the living marketing strategy doc at /var/www/html/systemprompt-web/reports/marketing/marketing-strategy-master.md. Never rewrites wholesale — writes diffs. Contains objectives, 30/60/90 targets, channel hypotheses in flight, dead hypotheses, winning tactics, and funnel snapshots. Load marketing-identity first.
Deterministically audit and optimise a published documentation page. Runs a 10-section quality audit, applies 5 rewrite rules for structure, terminology, link health, and completeness. Reads website analytics and GSC data per doc URL, produces a 75-point score delta (100 with analytics), commits changes, and updates the per-doc report. Load identity and brand-voice first.
Deterministically audit and optimise a published feature page on the register ladder: struggling-moment heroes with zero implementation jargon, mechanism evidence relocated to section bodies and dropdowns in the HashiCorp / Stripe / Tailscale register. Runs an 11-section quality audit plus register-ladder checks (hero jargon count, homepage narrative echo, struggling-moment opener, identifier-enumeration, established-not-indie voice) and applies 6 rewrite rules: claim verification, conversion clarity, brand discipline, and Technical-Marketing Synthesis (ten sub-checks: outcome headlines, jargon payoff scoped to sections/dropdowns, numbers with context, feature-to-outcome binding, narrative-vs-reference separation, skeptic test, named surfaces over coined metaphors, dropdown alignment, no Rust internals in narrative, dense sentences no filler). Penalises metaphor-stacking and marketing adjectives. Reads website analytics and GSC data per feature URL, produces a 90-point score delta (115 with analytics), commit
Deterministically audit and optimise a published guide. Runs a 14-section quality audit, applies 7 rewrite rules for value density, brand discipline, search-intent alignment, and CTR. Reads 28-day GSC query data per URL, produces a 100-point score delta across 11 dimensions, commits changes, and updates the per-guide report. Handles guides without GSC data. Load identity and brand-voice first.
Daily CRM and funnel measurement. Pulls GitHub Traffic API for systemprompt-core and systemprompt-template (14d retention, MUST run daily), website analytics via systemprompt CLI, and external feedback signals. Emits 1d/7d/31d funnel deltas and a dated report. Source of truth for every hypothesis metric.
Daily SEO briefing. Analyses content traffic, engagement, and search performance across published guides using internal analytics and Google Search Console. Generates actionable reports with S-### hypotheses. Designed for daily /loop. Load identity first.
| name | identity |
| description | The foundational source of truth for what systemprompt.io is, who it serves, and how it goes to market. Load this skill before any other content skill. Defines product identity, ICP, go-to-market strategy, competitive positioning, and messaging hierarchy. |
| metadata | {"version":"3.0.0","git_hash":"a5b0f4d"} |
The single source of truth for what systemprompt is, who it serves, and how it goes to market. Every other systemprompt skill must align with this document. Load it first, always.
One sentence: systemprompt.io is the only AI infrastructure you actually own.
The differentiation principle: The AI market is a sea of indistinguishable rented tools. systemprompt.io stands out on the one claim nobody else can make: you install it on your own systems, build on it, and keep it forever. Ownership is the lead message everywhere; control and governance are the first elaboration of what owning it gives you.
Full definition: systemprompt.io is AI governance infrastructure. It lets organizations own, standardize, and control how AI is used (Claude, Codex, Gemini, or in-house agents) across their teams, their departments, and (through white-label deployment) their customers. Provider support is verified in source: Anthropic, OpenAI, and Gemini providers plus a /v1 gateway that routes to any compatible upstream. Claude remains the best-supported client and the main SEO surface, but positioning is provider-neutral, Claude-aware: lead with governing every AI client, name Claude as one of them.
systemprompt is not a consumer app. It is not a prompt library. It is not an MCP server (though it includes one). It is infrastructure that other companies build on.
The product has four faces:
For enterprise security teams: A governance pipeline that integrates with existing security infrastructure. SIEM-compatible structured JSON audit logs, role-based access control with six tiers, end-to-end request tracing, and provable compliance for every AI agent interaction. Runs on-premises, air-gapped if needed, no data leaves the perimeter.
For mid-market engineering teams (SMEs): Standardized AI implementation across the organization. A centralised skill marketplace, shared knowledge base, usage analytics, and cost visibility. Every team member uses AI with the same knowledge, the same rules, the same processes, whichever client they run. Full observability of what AI you have, how it is used, and where it is used.
For individual users: A free tier that provides ownership and portability of your Claude plugins. Your skills, agents, and connectors stored securely in a third-party cloud, decoupled from any single ecosystem. Reusable, shareable, yours.
For white-label partners (SaaS companies): AI governance infrastructure deployed under their own brand. A custom-branded gateway (like the Dynapps Enterprise Intelligence Gateway) that lets SaaS companies offer AI governance to their own customers without building it themselves.
The common thread across all four: Ownership of your AI. You are firmly in control of what AI you have, how it is used, and where it is used, with full observability and permissions.
These are the concrete proof points that content skills should draw from when specificity is needed:
Every organization using Claude faces the same problem: AI adoption without governance is chaos. Different team members get different results. Knowledge disappears between sessions. There are no standards, no observability, no enforcement. The bigger the organization, the worse it gets.
Some try to build governance in-house. This is a trap. The AI landscape moves so fast that whatever you build today will need rewriting in three months. Anthropic ships new features, new plugin architectures, new capabilities constantly. Maintaining internal AI governance tooling means dedicating engineering resources to a problem that never stops moving.
systemprompt solves this permanently. Organizations get a governance layer that evolves with the ecosystem. When Anthropic ships something new, systemprompt supports it. When new governance requirements emerge, systemprompt handles them. The organization stays current without maintaining a single line of governance code.
The value proposition is ownership without the maintenance burden. You own the infrastructure; we keep it current.
systemprompt serves four audiences. ICP 1 is the content and credibility audience. ICP 2 is the primary sales and outreach target. This distinction matters: downstream marketing, CRM, and outreach skills must respect it.
Buyer: CISO, Security Engineering Lead, VP Security at companies of 1,000+ employees with dedicated security teams.
Entry point: Governance pipeline, SIEM integration, audit trails, secret detection.
The moment of pain: Security teams discover that AI agents are a blind spot in the SOC. Agents make tool calls, access external APIs, and process sensitive data with no centralised record. SIEM systems have no visibility. There is no policy enforcement on AI tool calls, no audit trail for compliance, and no secret protection. The CISO asks "what are our AI agents doing?" and nobody can answer.
What they say: "We need visibility into AI agent activity." "Our SIEM has a blind spot." "We need provable governance, not a dashboard."
Value frame: Defence, compliance, provable governance. systemprompt is a control, not a feature.
Vocabulary: governance pipeline, SIEM integration, RBAC, transport-layer governance, end-to-end request trace, audit trail, secret detection, air-gap, policy enforcement, provable governance.
Outreach rule: This is a content and credibility audience. Do not cold-email CISOs. Build authority through governance-focused content that security leaders find when they search for AI agent governance solutions.
Buyer: VP Engineering, Head of AI, CTO, Platform Engineering Lead at companies of 50 to 500 employees that have already standardised on Claude Code.
Entry point: Skill marketplace, plugin management, usage dashboard, cost visibility.
The moment of pain: A CTO realises that 30 people in their org are all using Claude differently, with different context, different quality, and no visibility into what is happening. Knowledge is siloed in individual developers. There is no centralised skill library. No one knows what AI costs per department. There is no way to enforce standards. They need governance, and they need it without building it.
What they say: "How do we standardise Claude Code across teams?" "I can't tell who's using it or what it costs." "We need shared skills and MCP servers." "Our security team is starting to ask about agent governance."
Value frame: Enablement, productivity, knowledge sharing, cost visibility. systemprompt turns fragmented individual AI usage into a governed, shared, measured capability.
Vocabulary: centralised skill library, usage analytics, cost attribution, role-based distribution, shared knowledge base, adoption metrics, governed by default.
Outreach rule: This is the primary sales and outreach target. Direct, relationship-driven engagement. The template hook ("clone the template, own the binary, see your team's AI usage by Friday") is calibrated for this audience.
SaaS companies whose customers are asking for AI governance capabilities. These companies face a build-vs-buy decision: dedicate engineering resources to building AI governance (and maintaining it as the landscape evolves), or partner with systemprompt and deploy a white-labelled solution.
The moment of pain: A SaaS company's customers start asking "can we govern how our teams use AI through your platform?" and the SaaS company realizes building this would consume their engineering team for months, and by the time they ship it, the AI landscape will have moved.
Individual Claude users who want ownership and portability of their plugins. These users matter for brand awareness and credibility, not for direct revenue. When a CTO Googles systemprompt after receiving a cold email, these users and their activity create the signal that this is real, actively used software.
systemprompt's primary go-to-market is direct, relationship-driven:
This solves the trust problem twice. The partner trusts the technology because they use it internally. Their customers trust it because it comes from a brand they already have a relationship with.
Content and SEO are supporting channels that build credibility and inbound discovery. They are not the primary growth engine. The primary motion is direct, relationship-driven outreach. Content exists to ensure that when a prospect researches systemprompt after a direct touch, they find authoritative, governance-focused material. Content also serves ICP 1 (enterprise security), where direct outreach is not appropriate but search-driven discovery is.
The website's primary job is enterprise credibility. When a CTO who has received a cold email visits systemprompt.io to vet the company before taking a meeting, the site must communicate "this is AI infrastructure you own" within 10 seconds. When a CISO searches for AI agent governance solutions, the site must demonstrate technical depth and security-first infrastructure.
The website's secondary job is supporting the free tier for individual users, which builds visible activity and brand familiarity.
The website must NOT look like a consumer product. The first impression must be enterprise-grade AI governance infrastructure.
The free tier is a demonstration environment. It proves the technology works. It creates visible user activity that signals real, active software. It gives individual users genuine value (ownership and portability of their plugins). But it is not the business model. It is a credibility engine.
The primary competitor is not another vendor. It is the internal engineering team that says "we can build this ourselves."
The argument against building in-house: The AI landscape evolves so rapidly that internal governance tooling becomes a maintenance burden that never ends. Anthropic ships new features, plugin architectures, and capabilities continuously. An in-house solution requires a dedicated team tracking every change, rewriting integrations, and maintaining compatibility. systemprompt absorbs that complexity. The organization gets governance that evolves with the ecosystem, maintained by a team whose entire focus is staying current with the AI landscape.
The one-line version: You could build it yourself. But by the time you ship it, you will need to rebuild it.
"systemprompt.io is the only AI infrastructure you actually own."
Use this as the anchor. Control of every AI client your organisation uses is the first elaboration of ownership; everything else follows from it.
To a CISO or security leader evaluating governance: "Every AI agent request traced end-to-end. SIEM-compatible audit logs. RBAC at the transport layer. One binary, your infrastructure, provable governance."
To a CTO evaluating for their org: "Standardize how your organization uses AI. Full observability, enforcement, and permissions across every team and every client."
To a SaaS company evaluating white-label: "Give your customers AI governance without building it yourself. Deploy under your own brand. We handle the infrastructure."
To an individual user: "Own your Claude plugins. Portable, shareable, securely stored, not locked to any ecosystem."
"System prompt" is a technical term in AI. It is the foundational instruction that controls how an AI behaves. The name is an asset: it signals to CTOs and technical leaders that this infrastructure operates at the foundational layer of AI control. It communicates authority and depth to the people who matter most (the primary ICP).
All content created using any systemprompt skill must align with this identity document. Specifically:
