- Home
- Skills
- Branding & Design
- Client Feedback Interpreter
Works with the AI Agents you already use
Client Feedback Interpreter
Clients describe feelings, not solutions, and a misread comment means a wasted revision round.
$19
Client Feedback Interpreter
Example session with this skill installed
My client sent these notes on the homepage draft. Turn them into change requests I can act on:
Thanks for the new version. Overall the structure works. Could we move the testimonials above the pricing? We need real photos of the team instead of the stock ones. The hero needs to pop more. Also, honestly, the footer feels off.
- Read your context and instructions
- Compiled the client feedback interpreter
Every sentence of the feedback is accounted for. Validator: PASS.
{
"schema_version": "1.0",
"language": "en",
"raw_feedback": "Thanks for the new version. Overall the structure works. Could we move the testimonials above the pricing? We need real photos of the team instead of the stock ones. The hero needs to pop more. Also, honestly, the footer feels off.",
"context": null,
"sentiment": "mixed",
"status": "needs_clarification",
"change_requests": [
{
"id": "FB-001",
"type": "structure",
"target": "testimonials",
"intent": "Move the testimonials section so it appears before the pricing section.",
"priority": "medium",
"confidence": 0.9,
"requires_clarification": false,
"clarification_question": null,
"possible_interpretations": [],
"requires_confirmation": false,
"fix_layer": "structure",
"raw_request": "Could we move the testimonials above the pricing?",
"structured_intent": {
"action": "change_order", "component": null, "section": "testimonials", "placement": "above",
"relative_to": "pricing", "size": null, "intensity": null, "appearance": null,
"new_text": null, "affected_color": null, "affected_cta": null, "affected_section": "testimonials",
"numeric_value": null, "basis": "explicit"
}
},
{
"id": "FB-002",
"type": "asset",
"target": "team",
"intent": "Replace the stock images in the team section with real photos of the team (the client must supply them).",
"priority": "medium",
"confidence": 0.9,
"requires_clarification": false,
"clarification_question": null,
"possible_interpretations": [],
"requires_confirmation": true,
"fix_layer": "assets",
"raw_request": "We need real photos of the team instead of the stock ones.",
"structured_intent": {
"action": "replace", "component": "image", "section": "team", "placement": null,
"relative_to": null, "size": null, "intensity": null, "appearance": null,
"new_text": null, "affected_color": null, "affected_cta": null, "affected_section": "team",
"numeric_value": null, "basis": "explicit"
}
},
{
"id": "FB-003",
"type": "visual_identity",
"target": "hero",
"intent": "Give the hero more visual impact. It is not yet clear whether through contrast/color, headline scale or imagery.",
"priority": "medium",
"confidence": 0.45,
"requires_clarification": true,
"clarification_question": "When you say the hero should \"pop\" more, do you mean stronger colors and contrast, a bigger headline, or a more striking image?",
"possible_interpretations": ["visual_identity", "layout", "asset"],
"requires_confirmation": false,
"fix_layer": "visual",
"raw_request": "The hero needs to pop more.",
"structured_intent": {
"action": "emphasize", "component": null, "section": "hero", "placement": null,
"relative_to": null, "size": null, "intensity": "more_visual_weight", "appearance": null,
"new_text": null, "affected_color": null, "affected_cta": null, "affected_section": "hero",
"numeric_value": null, "basis": "professional_inference"
}
},
{
"id": "FB-004",
"type": "unknown",
"target": "footer",
"intent": "The client is dissatisfied with the footer but does not say what is wrong.",
"priority": "low",
"confidence": 0.3,
"requires_clarification": true,
"clarification_question": "What feels off about the footer: the information it shows, how it looks, or its size?",
"possible_interpretations": ["copy", "visual_identity", "layout"],
"requires_confirmation": false,
"fix_layer": "unknown",
"raw_request": "Also, honestly, the footer feels off.",
"structured_intent": {
"action": "unknown", "component": null, "section": "footer", "placement": null,
"relative_to": null, "size": null, "intensity": null, "appearance": null,
"new_text": null, "affected_color": null, "affected_cta": null, "affected_section": "footer",
"numeric_value": null, "basis": "professional_inference"
}
}
],
"non_actionable_comments": [
{"text": "Thanks for the new version.", "reason": "courtesy"},
{"text": "Overall the structure works.", "reason": "praise"}
],
"clarifying_questions": [
"When you say the hero should \"pop\" more, do you mean stronger colors and contrast, a bigger headline, or a more striking image?",
"What feels off about the footer: the information it shows, how it looks, or its size?"
],
"traceability": [
{"fragment": "Thanks for the new version.", "destination": "non_actionable"},
{"fragment": "Overall the structure works.", "destination": "non_actionable"},
{"fragment": "Could we move the testimonials above the pricing?", "destination": "FB-001"},
{"fragment": "We need real photos of the team instead of the stock ones.", "destination": "FB-002"},
{"fragment": "The hero needs to pop more.", "destination": "FB-003"},
{"fragment": "Also, honestly, the footer feels off.", "destination": "FB-004"}
],
"summary": "4 change requests; 2 are ready and 2 need clarification (hero 'pop', footer) before the change plan can proceed."
}
Next steps: send the clarification questions to the client; apply the high-confidence changes now.
Connects securely to your tools. The creator never sees your data.
About this skill
by FS Studio
"Que respire más." "It needs to pop." "No me convence." This skill turns client comments into a change list you can act on, without guessing.
Client feedback is the most expensive part of a web project. Clients describe feelings, not solutions, and a misread comment means a wasted revision round. This skill reads feedback like a senior UI/UX designer and art director, and turns it into structured, traceable change requests:
- Perceptions become technical intent. "Quiero que respire más" becomes more air → change spacing; "se siente pesado" becomes less visual weight → simplify; "más premium" becomes premiumize. The lexicon covers Spanish and English.
- Literal vs interpreted, always explicit. Each request is marked
explicit(the client said it) orprofessional_inference(your professional reading). An interpretation is never presented as a confirmed requirement. - The client's words are preserved. Each request carries the exact fragment it came from. The validator rejects paraphrases, invented button texts, invented colors and invented numbers.
- Ambiguity becomes a question. Vague comments get a specific question you can forward to the client ("When you say 'pop', do you mean stronger colors, a bigger headline or a more striking image?"), and the plan is marked
needs_clarificationinstead of guessing. - Nothing is silently ignored. Every sentence of the feedback ends up somewhere: a change request, or a non-actionable comment (praise, courtesy). The validator checks this.
- Safe by default. Removals and replacements are marked
requires_confirmation. - Builder-agnostic. The intent vocabulary (24 actions, 27 components, 24 placements, 10 sizes, 10 intensities, 14 appearances) works the same for HTML, WordPress, Webflow, Shopify or Figma.
Use cases
- A client sends a WhatsApp message or email with comments on a preview.
- Meeting notes from a review call need to become a task list.
- An agency wants every feedback round documented the same way.
- A developer wants to feed structured change requests to another agent or a ticket system.
What you get
| File | Purpose |
|---|---|
SKILL.md | The skill instructions (4-step workflow plus quality gates) |
reference/interpretation-rules.md | Splitting rules, category precedence, closed vocabularies, perception lexicon (es/en), confidence, priority and confirmation rules |
reference/change-requests.schema.json | The JSON contract (v1.0) |
scripts/validate_feedback.py | Validator: schema, verbatim preservation, no invented values, clarification rules, full traceability |
examples/ | 3 complete cases: mixed feedback (es), agency notes (en), totally vague feedback (es) |
tests/ | 24 unit tests, including edge cases |
Example
Input:
El inicio lo veo muy apretado, quiero que respire más. El botón principal debería decir "Reservá tu clase". Sacaría la sección de preguntas frecuentes.
Output (excerpt):
| ID | Type | Literal / interpretation | Intent | Notes |
|---|---|---|---|---|
| FB-001 | visual_identity | interpretation | More whitespace in the hero (more_air, change_spacing) | |
| FB-002 | cta | literal | CTA text → "Reservá tu clase" | |
| FB-003 | remove | literal | Remove the FAQ section | requires confirmation |
Requirements
- An AI agent that supports Agent Skills (
SKILL.md). - Optional: Python 3.9+ for the validator. It uses the standard library only and makes no network calls.
Implementation status
| Feature | Status |
|---|---|
| Splitting feedback into fragments, with 10-category precedence | ✅ Included |
| Structured intent with closed vocabularies | ✅ Included |
| Perception lexicon (es/en) | ✅ Included |
Confidence, clarification questions, needs_clarification status | ✅ Included |
| Full traceability (nothing silently ignored) | ✅ Included, enforced by the validator |
| Confirmation flag for destructive changes | ✅ Included |
| Validator | ✅ Included, with 24 unit tests |
| Applying the changes to the site | ❌ Not included. The skill only interprets |
| Resolving conflicts between several stakeholders | ❌ Not included. Conflicts are recorded, not decided |
| Interpreting annotated screenshots or images | ❌ Not included. Text feedback only |
| Effort or cost estimation | ❌ Not included |
Compatibility and testing
Field-tested on 2026-11-03 with the final release. Every test gives the agent a normal user request without naming the skill. The output is checked with the skill's own validator plus case-specific expectations: correct classification, no invented data, the right verdict and so on.
| Environment | Status | Evidence |
|---|---|---|
| Claude Code (CLI 2.1.317) | ✅ Functionally tested | 8/8 cases passed. The skill was discovered and loaded automatically from .claude/skills/ in every run |
| OpenAI Codex CLI (0.164.0) | ✅ Functionally tested | 8/8 cases passed. The skill was discovered automatically from .agents/skills/ in every run |
Built on the open Agent Skills standard: a plain SKILL.md with name/description frontmatter, relative paths only and no tool-specific features.
What the 8 cases cover: 5 cases use inputs that are not in the shipped examples, so the agent cannot copy an example. The other 3 reproduce the shipped examples and only show that the examples can be reproduced. The cases include realistic inputs and edge cases: incomplete input, ambiguous input, unusual wording and mixed languages.
Validator and unit tests: these run anywhere with Python 3.9+ (python -m unittest discover -s tests).
AI agents are not deterministic. Two runs on the same input can phrase things differently or find slightly different sets of minor issues. The contract, the closed vocabularies and the validator keep the structure and the critical decisions consistent.
Pricing
US$19: a single license for your own and your clients' projects. It will also be part of the Web Proposal Pack for US$39.
License
FS Studio Commercial Skill License (single user). See LICENSE.txt.
- ✅ Use it in personal and commercial projects, including paid client work.
- ✅ The outputs you generate (blueprints, reviews, change requests) are yours.
- ❌ No redistribution, sharing or resale of the skill or its files, alone or bundled in another product.
The license supplements the Agensi Terms of Service, which prevail in case of conflict.
© 2026 Facundo Sicart · Published by FS Studio
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Branding & Design.
Does not apply the changes. Does not resolve conflicts between several client stakeholders; it records them. Text only: does not interpret annotated screenshots. No effort or cost estimates.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 14 days ago
Nothing to install. It runs in your AI agent.
- Optional
Python 3.9+
Python only runs the bundled validator, with the standard library.