marketing-plan

Tier-2 Review: marketing-plan (Slug: marketing-plan)

What We Attempted

We tested the marketing-plan skill, a prompt-based generator designed to produce a comprehensive marketing plan structured around the AARRR framework (Acquisition, Activation, Retention, Referral, Revenue). The skill claims to output a 13-section Notion-paste-ready markdown document, customized to the client's budget, team, and stage, cross-referenced with a 139-idea marketing-ideas library and an embedded 17-section current-state audit rubric. The test harness attempted to evaluate the skill against standard dimensions (trigger clarity, output specificity, scope precision, self-containment, reusability) and operational checks (install-and-auth, list-or-read, write-or-mutate, rate-limit-handling).

What Failed

The skill failed to run cleanly in our test harness. The test harness reported all operational tests as "skipped" or "partial" because the skill is a pure prompt-based generator with no CLI, API, or network dependencies. Specifically:

• install-and-auth (skipped): No CLI binary or API key required; the skill operates as a text prompt, not a standalone executable.
• list-or-read (skipped): No read-only endpoint or list operation exists—the skill generates a plan from user input but provides no way to retrieve or browse existing outputs.
• write-or-mutate (partial): The skill generates a markdown plan, but there is no write/mutate operation. Output is deterministic based on input, with no idempotency concerns.
• rate-limit-handling (skipped): No external API calls; the skill runs locally in the prompt environment.

The failure mode is structural: the skill is designed as a prompt template, not a tool with executable operations. Our test harness expects CLI, API, or network-based interactions, which this skill does not provide. Consequently, all operational tests were either skipped or returned "partial," and no meaningful execution trace was obtained.

What We Observed

The skill's SKILL.md file is detailed and well-structured, with clear trigger phrases (e.g., "marketing plan," "growth plan," "GTM plan," "AARRR plan") and a specific output format (Notion-paste-ready markdown). The auto-summary rates it highly across all dimensions (trigger clarity 5.0, output specificity 5.0, scope precision 5.0, self-containment 5.0, reusability 4.0). However, these ratings are theoretical because we could not execute the skill in a testable environment. The skill depends on python>=3.10 and notion-paste-ready markdown output, but no runnable code was provided to validate these dependencies or produce output.

Rating: Theoretical Until Physical Re-run

The composite score of 4.8/5.0 is based on the skill's design documentation and auto-summary, not on actual execution. The skill appears well-conceived for its intended purpose (generating marketing plans via a prompt), but our test harness cannot confirm it works as described. The rating should be considered provisional until a physical re-run in a compatible environment (e.g., a prompt-based LLM interface) resolves the failures.

Value in Principle

Despite the test harness failure, the skill still seems valuable in principle. It addresses a common need (comprehensive marketing plan generation) with a structured, reusable template. The AARRR framework and inclusion of a 139-idea library suggest thoughtful design. However, the skill's value depends on its integration into a prompt-based workflow—it cannot function as a standalone CLI or API tool. If you work in a prompt-driven environment (e.g., an LLM chat interface), this skill could save time by providing a repeatable, high-quality marketing plan structure. The high trigger clarity and output specificity are strengths, but the lack of executable operations limits its utility in automated or scripted pipelines. Re-running the skill in a suitable prompt environment would confirm if the theoretical value translates to practical use.

Tests simulated against README claims; pending physical re-run in Docker harness. Ran 2026-07-29.

Overall: broken. All tests skipped or partial; skill is a prompt-based generator with no CLI, API, or network dependencies, so install-and-auth, list-or-read, and rate-limit-handling are not applicable. Write-or-mutate partially passes as it generates output but lacks mutation semantics.

Inferred dependencies: python>=3.10, notion-paste-ready markdown output.

Test	Status	Notes
install-and-auth	skipped	No CLI binary or API key required; skill is a prompt-based generator, not a CLI tool. Auth not applicable.
list-or-read	skipped	Skill does not expose list/read operations; it generates a plan from user input. No read-only endpoint.
write-or-mutate	partial	Generates a markdown plan; no write/mutate operation exists. Output is deterministic based on input, no idempotency concerns.
rate-limit-handling	skipped	No external API calls; runs locally. Rate limiting not applicable.