file-save-protocol

Enforce persistent JSON and Markdown file saves with numeric prefixes.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/yjang-git/HoneyPot --skill file-save-protocol-yjang-git
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: file-save-protocol
Source: https://github.com/yjang-git/HoneyPot/tree/main/plugins/investments-portfolio/skills/file-save-protocol
Command: npx skills add https://github.com/yjang-git/HoneyPot --skill file-save-protocol-yjang-git

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Prevents data loss, context truncation, and AI hallucination by enforcing a strict file-saving workflow for analysis outputs so downstream agents and coordinators can reproduce and validate results.

Core Features & Use Cases

  • Mandatory persistence: Requires every analysis output to be written with the Write tool to a coordinator-provided output_path rather than returned only in prompts.
  • Dual-format storage: Enforces saving both machine-readable JSON and a human-readable Markdown summary with numeric filename prefixes for ordering and traceability.
  • Robust error handling & session resume: Defines explicit SUCCESS/FAIL responses with error codes (PATH_NOT_FOUND, PERMISSION_DENIED, DISK_FULL, UNKNOWN_ERROR) and includes session resume verification that checks for required JSON files before progressing.
  • Use Case: Orchestrating a DC pension portfolio analysis where multiple agents (index-fetcher, rate-analyst, sector-analyst, etc.) each produce JSON outputs and coordinated numbered markdown summaries consumed by downstream agents.

Quick Start

Use the Write tool to save the analysis_result as portfolios/{session_folder}/rate-analysis.json and also write a numbered markdown summary 00-rate-analysis.md into the same output_path.

Frequently Asked Questions about file-save-protocol

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I save agent analysis outputs to persistent files and prevent hallucination?

To prevent hallucination and context truncation in agent analysis outputs, save results to persistent files using a strict workflow that enforces writing both JSON data and markdown summaries with the Write tool, rather than returning them only in prompts.

Why does my multi-agent portfolio analysis lose data during context truncation?

Multi-agent portfolio analysis loses data during context truncation when outputs are returned only in prompts instead of saved to persistent files, a problem solved by enforcing a strict file-saving workflow that writes both JSON and markdown to a coordinator-provided output path.

How to write both JSON and markdown files for reproducible analysis pipelines?

Write reproducible analysis pipeline outputs by using the Write tool to save machine-readable JSON data and a human-readable markdown summary into a coordinator-provided output path, applying numeric filename prefixes for ordering and traceability.

Does this file-saving protocol support session resume verification for downstream agents?

Yes, session resume verification is supported by checking for required JSON files before progressing, and the protocol returns structured SUCCESS or FAIL responses with error codes like PATH_NOT_FOUND, PERMISSION_DENIED, or DISK_FULL for robust error handling.

What is the best way to orchestrate multi-agent portfolio analysis with persistent file outputs?

The best way to orchestrate multi-agent portfolio analysis is enforcing a file-save protocol where each agent writes numbered markdown summaries and JSON outputs to a shared session folder, allowing downstream agents and coordinators to reproduce and validate results.

How do I handle file save errors when writing analysis outputs to an output path?

Handle file save errors during analysis output persistence by returning structured FAIL responses with specific error codes such as PATH_NOT_FOUND, PERMISSION_DENIED, DISK_FULL, or UNKNOWN_ERROR, ensuring the coordinator can manage failures and resume sessions.