PROMPT TEMPLATE MED V0.0.1
# Meta-Structure Definition - Medium Complexity Tasks v0.8.0
**Date:** [YYYY-MM-DD]
**Time:** [HH:MM UTC+/-Offset]
## 1. Overall Purpose
This document defines a structured template for prompts intended to guide an LLM agent in executing moderately complex, sequential tasks. It balances the need for clear instructions and verification with reduced overhead compared to the full autonomous execution template, making it suitable for feature implementation, configuration setups, or significant refactoring tasks.
## 2. Core Execution Principles
* **Sequential Execution:** Tasks and Steps MUST be executed in the listed order. Each item MUST begin with `- [ ]` and be marked complete `- [x]` before proceeding to the next.
* **Task-Level Verification:** Each Task includes a `Task Verification` section outlining expected outcomes and checks. The agent MUST perform these checks upon completing the Task's steps.
* **Error Handling:** If `Task Verification` fails or a significant error occurs during a step:
* Attempt **one** cycle of identifying the issue, correcting the relevant step(s), and re-running the verification.
* If verification still fails after the correction attempt, STOP execution and report the error and the state of the task. Do not proceed.
* **Autonomous Operation & Reporting:** Execution should be autonomous. Report only upon successful completion of all tasks or when an unrecoverable error (per the rule above) occurs. Internal logging should be minimal unless specified.
* **Quality & Best Practices:** Adhere to general software development best practices (e.g., clarity, modularity, security considerations) and any project-specific guidelines mentioned within the prompt or readily available (e.g., a `CONTRIBUTING.md` or style guide). Explicit linked rule adherence like in the Full template is *not* required unless specifically stated for a critical step.
## 3. Prompt Template Structure
* **A. Overall Formatting:**
* Enclose the entire prompt in ```markdown ... ```. *(Note: This instruction refers to how the prompt *instance* should be formatted when used, not this template definition itself)*
* All Task and Step items MUST begin with `- [ ]`.
* **B. Subject Line:**
* Format: `Subject: [ACTION-ORIENTED TITLE for Medium Task]`
* **C. Directive Section:**
* Clearly state the overall goal of the prompt.
* Mention the requirement for sequential execution, completion marking (`- [x]`), adherence to best practices, the task-level verification process, and the reporting conditions (completion or error after retry).
* Example: "Execute the following tasks sequentially to [overall goal]. Mark checkboxes upon completion. Adhere to standard coding practices. Verify each task using the provided checks, attempting one correction cycle if needed. Report final status or any unrecoverable errors."
* **D. Execution Blocks:**
* Uses a `Task -> Step` hierarchy. Markdown headings/lists. Separated by `---` between Tasks is recommended for clarity.
* **Every Task and Step MUST start with `- [ ]`.**
* **D.1. Task Block (`- [ ] **Task X: Name**`)**
* Content: `* **Objective:**`, a list of Step items, and `* **Task Verification:**`.
* **D.2. Step (`- [ ] * **Step X.Y [(Optional Guideline/Best Practice)]:** Action`)**
* Content: Begins with an explicit **action verb**. Describes a single, logical action. May include optional, non-linked references to key guidelines or best practices like `(Best Practice: Modularity)` or `(Guideline: Use Config File)` if helpful.
* **D.3. Task Verification (within Task Block)**
* Content: Defines clear, checkable conditions for task success and the method to verify them. This combines the intent of `Internal Success Criteria` and `Internal Verification Method` from the Full template but is less formal. It MUST confirm that the steps were completed correctly and align with the Task Objective and mentioned best practices.
* Example: `* **Task Verification:** Check that the file 'src/new_module.py' exists and contains the 'NewClass' definition. Confirm basic linting passes on the new file. Verify the configuration file 'config/settings.yaml' includes the 'new_module_enabled: true' setting.`
* **E. Final Instruction:**
* Explicitly command the start of execution.
* Example: `Begin execution starting with Task 1, Step 1.1. Follow all instructions sequentially.`
* **F. Contextual Footer:**
* Optional: Include generation timestamp or other relevant metadata.
* `*(Generated: [YYYY-MM-DD HH:MM UTC+/-Offset]. Location: [Location])*`
## 4. Example Snippet (Illustrates how to *use* the template)when to use it
Community prompt sourced from the open-source GitHub repo justinlietz93/Perfect_Prompts (MIT). A "PROMPT TEMPLATE MED V0.0.1" style prompt — adapt the placeholders and specifics to your task. Imported as-is and not independently retested here, so check the output before relying on it.
tags
productivitycommunitydeveloper
source
justinlietz93/Perfect_Prompts · MIT
more in Productivity
Productivity✓ tested
Summarize a doc into decisions & actions
chief of staff who extracts what to DO, not just what was said
Productivity✓ tested
Draft a reply to a hard email
calm, direct communicator who de-escalates without caving
Productivity✓ tested
Turn a brain-dump into a weekly plan
planning coach who protects your focus, not just your calendar