- Home
- Skills
- Prompt & Skill Engineering
- Reverse Engineer Any Template Agent
More screenshots
Works with the AI Agents you already use
Reverse Engineer Any Template Agent
Instead of merely guessing what the original prompt might have looked like, the agent performs structured reverse engineering.
$9.99
Reverse Engineer Any Template Agent
Example session with this skill installed
I lost the prompt that generated my product descriptions, but I still have three examples and the inputs I used.
PAIR A
Input
Product = "Lavender Sleep Candle"
Audience = "women interested in relaxation and self-care"
Output
"Turn your evening into a calming ritual with the Lavender Sleep Candle, a soothing home fragrance designed for peaceful nights. Its gentle lavender-inspired aroma creates a warm, comforting atmosphere for reading, journaling, meditation, or winding down before bed. A thoughtful choice for self-care routines, cozy bedrooms, and relaxing gifts."
PAIR B
Input
Product = "Citrus Focus Candle"
Audience = "remote workers and students"
Output
"Refresh your workspace with the Citrus Focus Candle, an energizing home fragrance designed for productive mornings and focused work sessions. Its bright citrus-inspired aroma creates an uplifting atmosphere for studying, planning, writing, or completing demanding tasks. A practical choice for home offices, study spaces, and productivity-themed gifts."
PAIR C
Input
Product = "Vanilla Cozy Candle"
Audience = "people who enjoy cozy home decor"
Output
"Make your space feel instantly warmer with the Vanilla Cozy Candle, a comforting home fragrance designed for slow evenings and inviting interiors. Its soft vanilla-inspired aroma creates a cozy atmosphere for reading, relaxing, entertaining, or enjoying quiet time at home. A charming choice for bedrooms, living spaces, and thoughtful housewarming gifts."
Reconstruct the reusable prompt behind these examples.
I want
- Fixed instructions
- Variable-driven content
- Derived logic
- Negative constraints
- Reconstructed imperative prompt
- PromptBase-style INPUTS block using {{double_curly}} variables
- Confidence level for every inferred rule
- Validation tests
Do not claim that you recovered the exact original prompt.
- Read your context and instructions
- Compiled the reverse engineer any
Reconstruction Summary
The three examples strongly support a reusable product-description template with two direct buyer-controlled variables:
- {{product_name}}
- {{target_audience}}
The product's implied scent, use cases, atmosphere, and gift context appear to be derived automatically from the product concept rather than supplied as separate inputs.
Overall Reconstruction Confidence: HIGH
Generalization Scope: MODERATE
The template is strongly supported for lifestyle/home-fragrance products. Broader product categories should be tested separately.
Observed Invariants
| Rule | Evidence | Confidence |
|---|---|---|
| Output is one paragraph | All three examples | HIGH |
| Description contains three sentences | All three examples | HIGH |
| Product name appears in sentence one | All three examples | HIGH |
| Sentence one defines the product and primary benefit | All three examples | HIGH |
| Sentence two expands atmosphere and use cases | All three examples | HIGH |
| Sentence three positions suitable contexts or gifting | All three examples | HIGH |
| Tone is warm, polished, benefit-oriented, and non-technical | All three examples | HIGH |
| No headings or bullet lists appear in the generated description | All three examples | HIGH |
Variable Map
{{product_name}}
Direct variable.
Examples
- Lavender Sleep Candle
- Citrus Focus Candle
- Vanilla Cozy Candle
Observed Effect
Controls the named product and influences the semantic theme of the entire description.
Confidence
VERIFIED FROM INPUT DIFFERENCES
{{target_audience}}
Direct variable.
Observed Effect
Influences use cases, setting, emotional framing, and likely gift context without being repeated literally.
Transformation
Audience adaptation.
Confidence
HIGH
Derived Logic
R-01 — Infer Product Mood
Rule
Infer an emotional or functional mood from {{product_name}}.
Examples
Lavender Sleep → calming
Citrus Focus → energizing
Vanilla Cozy → comforting
Confidence
HIGH
Validation Test
Use {{product_name}} = "Eucalyptus Refresh Candle" and verify that the mood becomes fresh/revitalizing rather than copying "calming."
R-02 — Generate Contextually Appropriate Use Cases
Rule
Add several activities naturally associated with the product concept and {{target_audience}}.
Confidence
HIGH
Evidence
Reading, journaling, meditation, studying, planning, writing, entertaining, and relaxing change appropriately across examples.
R-03 — Generate Appropriate Placement / Gift Context
Rule
End with suitable environments, lifestyle contexts, or gifting situations.
Confidence
HIGH
Negative Constraints
Likely rules
- Do not use bullet points.
- Do not include headings.
- Do not use exaggerated claims.
- Do not invent technical product specifications.
- Do not repeat the target audience mechanically.
- Do not use aggressive sales language.
Confidence
MEDIUM-HIGH
These constraints are inferred from consistent output style rather than directly supplied instructions.
Reconstructed Prompt
INPUTS
{{product_name}} = The product to describe.
Example: "Lavender Sleep Candle"
Required.
{{target_audience}} = The primary audience or lifestyle group the description should resonate with.
Example: "women interested in relaxation and self-care"
Required.
PROMPT
Write one polished three-sentence product description for {{product_name}}, tailored naturally to {{target_audience}}.
In the first sentence, introduce the product by name, identify what kind of product it is, and communicate its primary emotional or functional benefit.
In the second sentence, infer details naturally associated with the product concept and describe the atmosphere, experience, or practical use cases it supports. Include several contextually relevant activities without mechanically repeating the audience input.
In the third sentence, identify appropriate environments, lifestyle situations, or gifting contexts where the product would fit naturally.
Use warm, premium, benefit-oriented language that feels commercially polished but not exaggerated. Adapt the mood and vocabulary intelligently to the meaning of {{product_name}} and the needs of {{target_audience}}.
Do not invent technical specifications, ingredients, certifications, performance guarantees, or factual product claims that were not supplied. Do not use headings, bullet points, labels, or additional commentary.
Return only the final three-sentence paragraph.
Validation Matrix
| Test | Input Change | Expected Result |
|---|---|---|
| Golden Example | Lavender Sleep Candle | Calm/self-care framing |
| Variable Test | Eucalyptus Refresh Candle | Fresh/revitalizing framing |
| Audience Test | Gift buyers | Stronger gifting context |
| Distant Variation | Cinnamon Holiday Candle | Seasonal/cozy framing |
| Format Test | Any input | Exactly one paragraph, three sentences |
| Constraint Test | Product name implies medical benefit | Avoid unsupported medical claims |
Unresolved Ambiguity
The examples do not prove whether "three sentences" was explicitly required in the original prompt or emerged from a broader instruction such as "write a short product description."
Confidence that the final template should preserve three sentences:
HIGH
Confidence that the lost original explicitly contained the words "three sentences":
LOW
Final Assessment
This reconstructed candidate explains the supplied examples with high confidence while keeping the number of buyer-controlled variables low. It is suitable for PromptBase testing as a reusable lifestyle-product description template.
It should be validated on at least three additional product categories before being marketed as broadly universal.
Connects securely to your tools. The creator never sees your data.
About this skill
Reverse Engineer Any Template Agent is a premium prompt-forensics, template reconstruction, and reusable-instruction engineering agent designed for situations where a user has a great AI-generated result but has lost, forgotten, or never properly documented the prompt that created it.
Instead of merely guessing what the original prompt might have looked like, the agent performs structured reverse engineering.
The user can provide
Example Output Known Input Values Multiple Input / Output Pairs Partial Memory of the Original Prompt PromptBase Examples Image Prompt Results SEO Descriptions Product Descriptions Structured Reports JSON Outputs Lesson Plans Prompt Packs Agent Outputs Writing Templates Design Prompts Generated Documents
The agent then works backward using the following model
Example Inputs + Example Outputs
- Evidence Inventory
- Difference Analysis
- Invariant Detection
- Input-to-Output Mapping
- Variable Detection
- Fixed-Instruction Detection
- Derived-Logic Inference
- Constraint Reconstruction
- Formatting Reconstruction
- Prompt Reconstruction
- PromptBase Variable Design
- Confidence Scoring
- Validation Testing
The core principle is
Differences reveal variables.
Similarities reveal fixed instructions.
This makes the skill especially powerful when several examples are available.
For example, if three generated image prompts differ only in:
Shop Type Shop-Specific Objects
while preserving
One Storefront Front View Watercolor Style White Background Pastel Palette No People No Neighboring Buildings
the agent can infer that the first group represents variable-driven content while the second group likely represents fixed instructions.
The reconstructed template can then expose
{{shop_type}}
and
{{shop_specific_objects}}
while keeping the visual architecture fixed.
The agent does not claim to recover inaccessible prompts verbatim.
A generated result does not uniquely identify the exact original instruction because multiple prompts can produce similar outputs.
The agent therefore treats reconstruction as
A best-fit template hypothesis
rather than
Proof of the exact original prompt
Every reconstructed rule can be labeled as
Directly supplied
Observed
Verified by user
High-confidence inference
Medium-confidence inference
Low-confidence inference
Unknown
This makes the reconstruction transparent and testable.
The skill supports several operating modes.
Single-example reconstruction
The user provides
One Example Output Known Inputs
The agent produces the strongest possible candidate template but explicitly identifies ambiguity caused by the limited sample.
Multi-example differential reconstruction
The preferred mode.
The user provides
Input A Output A
Input B Output B
Input C Output C
The agent compares all examples systematically and determines:
What Changed What Stayed Constant Which Changes Correspond to Inputs Which Details Were Derived Automatically Which Rules Were Probably Fixed
Promptbase reconstruction
Designed specifically for prompt sellers.
The agent produces
Reconstructed Prompt PromptBase-Style INPUTS Block {{double_curly}} Variables Variable Descriptions Required / Optional Status Example Values Fallback Behavior Validation Tests
Image prompt reconstruction
The agent analyzes
Subject Object Count Composition Camera / View Background Medium Visual Style Palette Lighting Texture Typography Negative Constraints Aspect Ratio Subject-Specific Details
Writing template reconstruction
The agent analyzes
Audience Tone Structure Opening Pattern Paragraph Structure
Cta
SEO Requirements Length Keyword Placement Formatting Negative Constraints
Structured output reconstruction
For
Json
Yaml
Reports Tables Decision Systems Checklists
The agent reconstructs
Required Keys Optional Keys Types Enumerations Nesting Ordering Output Contracts
Agent workflow reconstruction
When an output appears to come from a multi-step process, the agent can infer:
Inputs Validation Analysis Stages Decision Gates Audit Rules Warnings Output States Edge Cases Final Deliverables
The agent maintains a Rule Ledger.
Every important inferred rule can include
Rule ID Rule Rule Type Evidence Confidence Validation Test
Example
R-01
Rule
Generate exactly one isolated subject.
Type
Count / composition
Evidence
All six supplied examples contain exactly one primary subject.
Confidence
High
Test
Use a subject that would normally be represented as a group and verify that the template still generates one primary subject.
The agent classifies template logic into categories such as
Task rule
Fixed content rule
Variable rule
Derived variable rule
Optional variable rule
Conditional rule
Count rule
Order rule
Length rule
Style rule
Tone rule
Format rule
Negative constraint
Quality rule
Platform rule
Variable detection
The agent identifies which parts of the result are controlled directly by inputs.
Examples
{{subject}} {{industry}} {{target_audience}} {{product_category}} {{headline_text}} {{style}} {{color_palette}}
Derived variable detection
Some output content may not correspond to a separate user input.
Example
Input
{{shop_type}} = bakery
Output also contains
Bread Baskets Rolling Pins Pastries
The agent may infer
Add several visually appropriate accessories naturally associated with {{shop_type}}.
This is preferable to creating unnecessary variables when the AI can derive the content reliably.
Fixed-instruction detection
The agent identifies rules that remain stable across examples.
Examples
Use a White Background Create Exactly One Subject Use a Professional Tone Return JSON Only Keep the Subject Centered Generate Exactly 10 Variations Do Not Number the Outputs
Negative-constraint detection
The agent actively searches for hidden prohibitions.
Examples
No People No Watermark No Background Scenery No Additional Commentary No Numbering No Duplicate Subjects No YAML Frontmatter No Explanations
These rules are often critical to reproducing the original result.
Format reconstruction
The agent can infer
Markdown Plain Text
Json
Yaml
Tables Code Blocks One Code Block Separate Code Blocks Exact Label Order Exact Number of Sections
Count reconstruction
If every example consistently produces
10 Prompts
the agent can infer
Generate exactly 10 prompts.
However, count rules are not invented without evidence.
Length reconstruction
The agent can identify likely
Character Limits Word Limits Sentence Limits
when supported by examples, user memory, or marketplace constraints.
Style reconstruction
The agent examines
Vocabulary Sentence Length Formality Emotional Tone Technicality Visual Adjectives Rhetorical Style Commercial Language Educational Framing
Possible inferred styles include
Professional Premium Whimsical Analytical Academic Persuasive Technical Minimalist Playful Luxury Conversational
The agent prefers explicit imperative instructions over vague persona inflation.
Instead of relying on
"You are the world's greatest copywriter."
it reconstructs operational instructions that explain the actual result.
Promptbase variable design
The agent uses
{{double_curly}}
slots with descriptive names.
Good
{{target_audience}} {{product_category}} {{headline_text}} {{secondary_objects}}
Avoid
{{x}} {{var1}} {{thing}}
The agent also prevents excessive variable exposure.
A template should expose the smallest useful number of buyer-controlled variables.
More variables provide more control but increase buyer friction.
Fewer variables improve usability but require more model interpretation.
The agent can therefore recommend whether a detail should be:
Fixed
Variable
Derived automatically
Optional inputs
Optional variables can include explicit fallback behavior.
Example
If {{color_palette}} is blank, choose a palette appropriate to {{subject}} and {{visual_style}}.
Generalization analysis
The agent evaluates whether the reconstructed template is
Narrow
Moderate
Broad
Unknown
A template reconstructed only from watercolor flower shops should not automatically be described as universal.
It can also report
Generalization risk
when a buyer tries to apply the template outside the observed domain.
Competing hypotheses
When two explanations fit the evidence, the agent does not arbitrarily choose one.
Example
Hypothesis A
The pastel palette was explicitly fixed.
Hypothesis B
The AI selected pastel colors because every example had a cozy subject.
The agent can then propose a discriminating test.
Example
Use
{{subject}} = industrial steel factory
If the output remains pastel
Supports Hypothesis A.
If the palette changes
Supports Hypothesis B.
Validation engine
Every important inferred rule should be testable.
Supported validation approaches include
Golden Example Test One-Variable-at-a-Time Test Close Variation Test Distant Variation Test Negative Constraint Test Count Test Format Test Length Test Optional Input Test Ablation Test Out-of-Domain Test Edge-Case Test
Golden example test
The reconstructed template is run conceptually against the original known input.
It should reproduce
Structure Style Constraints Variable Behavior Major Output Characteristics
Exact wording is not required.
One-variable-at-a-time testing
Only one input is changed.
This allows the user to observe whether the correct output dimension changes.
Ablation testing
One inferred rule is removed.
If the result remains equally faithful
the rule may not be necessary.
Variable-leak detection
The agent audits reconstructed prompts for example-specific values that accidentally remained hardcoded.
Example
Incorrect
Include a rose.
Correct
Include {{subject}}.
Possible result
Rta-014 variable_leak
Overconstraint detection
The reconstructed prompt contains unsupported instructions that unnecessarily reduce variation.
Underconstraint detection
The reconstructed prompt fails to preserve an important repeated characteristic.
Mixed-template detection
If several examples appear to come from different original prompts:
the agent can flag
Possible mixed template set
Possible evidence includes
Conflicting Output Structures Incompatible Fixed Rules Radically Different Tone Different Variable Mapping
Manual-edit detection
If one output differs materially from the rest
the agent can flag
Manual_edit_suspected
because the final result may contain human editing that was not caused by the original prompt.
Post-processing detection
The agent can also consider whether the final result passed through:
Formatter Code Processor Spreadsheet Formula Image Tool Human Editor Marketplace Renderer
The goal is to avoid incorrectly attributing every visible characteristic to the prompt itself.
Model and platform sensitivity
The same prompt can behave differently across
ChatGPT Claude Gemini Image Models Custom Agents Future Model Versions
The agent can rate prompt portability as
High
Medium
Low
and identify behaviors that depend on platform-specific defaults.
The reconstructed prompt is written as a clean imperative instruction.
Recommended architecture
Task
- Input Definitions
- Fixed Rules
- Variable Handling
- Derived Logic
- Style Rules
- Negative Constraints
- Output Format
- Quality Check
For PromptBase sellers, the final product can include
Inputs
{{subject}} = The primary subject. Example: "a cozy village bakery" Required.
{{secondary_details}} = Optional details to integrate. Example: "bread baskets, pastries, rolling pins" Optional.
Prompt
[Complete reconstructed reusable instruction]
The agent can also produce
Filled Example Validation Matrix Rule Ledger Competing Hypotheses Unresolved Ambiguities Generalization Rating Prompt Portability Optimized Marketplace Version
The commercial value of the skill is that one successful result can become a reproducible product.
Instead of manually trying to remember
"What did I type to get this?"
the user receives
A Structured Template + Reusable Variables + Evidence-Based Rules + Validation Tests + Confidence Levels
This makes the recovered template suitable for
PromptBase Internal Prompt Libraries AI Workflow Documentation Agent Skills Reusable Team Templates Prompt Productization Testing Versioning Marketplace Sales
The defining standard is
Can a new buyer provide different input values and reproduce the same underlying behavior without relying on the original example?
If yes, the reconstruction has succeeded.
A generated output does not uniquely identify the exact prompt that created it.
Different prompts can produce very similar results.
Therefore reconstruction is probabilistic rather than forensic proof.
Single-example reconstruction is substantially less reliable than reconstruction based on multiple aligned input/output pairs.
Outputs may also reflect factors outside the prompt, including:
Model Randomness
Model Version
Platform Defaults
Manual Editing
Post-Processing
Formatting Tools
Image Generation Defaults
External Scripts
Marketplace Rendering
Some output characteristics may therefore not originate from the prompt itself.
A repeated pattern does not always prove that the original prompt explicitly required it.
For example:
Three examples containing exactly three sentences
can strongly justify preserving three sentences in the reconstructed template, but it does not prove the original prompt literally said:
"Write exactly three sentences."
Variable detection can also become ambiguous when multiple inputs change simultaneously.
The strongest differential analysis changes one input at a time.
Derived logic may be difficult to distinguish from hidden variables.
For example:
A bakery output containing pastries
could result from:
A specific pastries variable
or:
An instruction to add contextually appropriate bakery details.
Additional examples help distinguish these hypotheses.
Generalization should not be assumed.
A template reconstructed from a narrow set of examples can fail when applied to unrelated domains.
The same reconstructed prompt can also behave differently across:
ChatGPT Models
Claude
Gemini
Image Models
Custom Agents
Future Model Versions
The skill cannot extract inaccessible private system prompts, developer instructions, hidden chain-of-thought, internal policies, or secret backend configuration.
Its purpose is to reconstruct a reusable external template from observable and authorized evidence.
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 any payment.
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 2 months ago
Permissions
File Scopes
Use Read Files to inspect authorized example outputs, known input values, old prompt files, PromptBase assets, generated templates, screenshots, structured reports, JSON outputs, user notes, and existing reconstructed prompts. Use Write Files only when the user requests creation or updating of derived artifacts such as: Reconstructed Prompts PromptBase Templates INPUTS Blocks Variable Maps Rule Ledgers Validation Matrices Prompt Audits Test Cases Golden Examples Template Documentation Marketplace-Ready Prompt Files Use Browser only when current public information is genuinely required and authorized. Template reconstruction normally depends primarily on the user's own examples rather than external information. Always distinguish: DIRECTLY SUPPLIED Information provided explicitly by the user. OBSERVED A property directly visible in the examples. VERIFIED BY USER An inferred rule the user later confirms. HIGH-CONFIDENCE INFERENCE Strongly supported by multiple examples. MEDIUM-CONFIDENCE INFERENCE Plausible but not uniquely determined. LOW-CONFIDENCE INFERENCE Weakly supported or dependent on a single example. UNKNOWN Cannot be established. Do not claim: Exact Original Prompt Recovered unless the original prompt text itself has been supplied. Prefer: Best-Fit Reconstructed Template or: Candidate Template Explaining the Observed Examples When several examples are provided: Differences Reveal Variables. Similarities Reveal Fixed Instructions. For every known input, trace: Where It Appears How It Changes the Output Which Secondary Content It Influences Do not create unnecessary variables. Where the model can reliably derive a detail from another variable, consider representing it as derived logic instead. Use PromptBase-style: {{double_curly}} variables when marketplace-ready reconstruction is requested. Use descriptive variable names. Every important inferred rule should include: Evidence Confidence Validation Test Actively check for: Variable Leakage Overconstraint Underconstraint Mixed Template Sets Manual Editing Post-Processing Platform Dependence Generalization Risk The final template should reproduce behavior rather than merely memorize the example. A strong reconstruction should work with new input values while preserving the same underlying structure, style, constraints, and transformation logic.