Client Proposal Generator for Marketing Services
Turns discovery-call notes into a client-ready marketing proposal with tiered pricing, ROI projections, and follow-up sequences — via one Claude Code command.
Pure text-generation skill: no script execution, no network calls, no sensitive-data instructions; low overreach risk. However, no statement on how client data is handled, no confirmation mechanism or rollback guidance, and reliance on repo-level install.sh (not reviewed here); least-privilege and data-flow disclosure are incomplete.
Internally consistent instructions, clear trigger and output format (CLIENT-PROPOSAL.md); but edge cases (user withholds info, missing/malformed audit data) are only partially covered, no tests and no designed failure feedback. Static review caps this score.
Audience (agencies/consultants) and scenario (post-discovery proposal) are clear, trigger '/market proposal' is explicit, integration with /market audit and /market competitors is well defined; but capability boundaries and non-fit ranges are undeclared, and there is no Chinese-language support or mainland-China environment consideration.
Well-structured progressive disclosure (inputs → framework → templates → principles); repo has MIT license and uninstall script. But the skill itself has no versioning, changelog, known-limitation disclosure or FAQ; maintenance ownership and update path unclear.
As a proposal template it has real value: complete section structure, concrete scripting frameworks, produces directly editable Markdown; but output is template-filling with model-generated ROI figures (fabrication risk), and marginal advantage over existing proposal templates is unevidenced. Usability of final output unverifiable statically.
Author claims only: 'data-backed proposals close at 2-3x the rate', '40-60% increase in qualified leads' have no cited sources; no tests, no sample outputs, no third-party execution evidence.
- ROI projections are model-generated; verify all figures manually before sending to clients to avoid fabricated performance promises
- Effectiveness claims ('2-3x close rate') are unsourced marketing language
- No versioning or changelog; behavior may change without notice
- No Chinese-language support; localization required for Chinese-speaking clients
- Publisher identity unverified; maintenance continuity uncertain
What does this skill do, and when should you use it?
This is one of the 15 skills in the AI Marketing Suite (located at skills/market-proposal/SKILL.md), dedicated to producing proposals for prospective marketing clients. It gathers client context and service scope, then builds a complete document: executive summary, situation analysis, phased strategy, scope of work, timeline, three-tier pricing, ROI projection, team, case studies, and next steps. Output is saved as CLIENT-PROPOSAL.md. It also ships a 10-question discovery-call framework, a 21-day post-send follow-up email cadence, and objection-handling scripts. The skill is pure prompt instructions with no scripts; if a prior /market audit exists, its findings are folded into the situation analysis automatically.
Reads user-supplied client details (company, pain points, goals, budget range, decision timeline) and the provider's service offering; automatically incorporates prior /market audit or /market competitors results into the Situation Analysis when present; generates an 11-section proposal saved as CLIENT-PROPOSAL.md including a Good-Better-Best pricing table (Growth/Accelerate/Dominate), an ROI calculation framework, a milestone timeline, and a terms-and-conditions checklist; additionally supplies a 10-question discovery framework, a Day 0–Day 21 follow-up email sequence, and response frameworks for eight common objections.
- An agency or freelance consultant who has finished a discovery call and needs to formalize the engagement into a professional proposal quickly.
- A new marketing consultant without a proposal template who wants proven structure — pricing anchoring, tiered options, ROI framing — for their first proposal.
- An agency that already ran /market audit on a prospect's site and wants the audit findings baked into the proposal to raise close rates.
- A sales team that wants pre-written responses to objections like 'too expensive' or 'we can do this in-house'.
- A solopreneur who sends proposals but lacks follow-up discipline and wants a Day 2 to Day 21 email cadence.
What are this skill's strengths and limitations?
- Complete, professional proposal structure: 11 sections from executive summary to T&C essentials, with fill-in template text.
- Concrete pricing guidance: a three-tier pricing table template, pricing psychology tips, and five pricing models with typical market ranges.
- Goes beyond document generation: the discovery question list, follow-up sequence, and objection table are directly usable sales assets.
- Reuses /market audit results automatically when available, making the proposal data-backed.
- ROI projections are entirely template-driven from user input; the skill does not validate numbers, so over-promising risk sits with the user.
- Templates default to English and Western market conventions (USD pricing, Net 15/30 terms) and need adapting elsewhere.
- No built-in PDF export — PDF capability belongs to sibling skills in the suite (and requires optional reportlab); this skill outputs Markdown only.
- No tests or sample outputs accompany the SKILL.md; output quality depends on how well the model follows the long instruction set.
How do you install this skill?
Install the full suite first: curl -fsSL https://raw.githubusercontent.com/zubair-trabzada/ai-marketing-claude/main/install.sh | bash, or clone the repo and run ./install.sh. The installer places skills under ~/.claude/skills/. The repo does not document installing this single sub-skill in isolation — check install.sh if that's needed.
How do you use this skill?
In Claude Code, type /market proposal or /market proposal <client name>. When prompted, provide the client's name, industry, pain points, goals, budget range, plus the services and engagement model you're proposing. If no discovery call has happened yet, ask the skill for its built-in 10-question list. The generated proposal is written to CLIENT-PROPOSAL.md. For a data-backed Situation Analysis, run /market audit <url> in the same environment beforehand.