- Home
- Skills
- Agents & Orchestration
- paper trail
paper trail
Turns messy multi-platform AI chats into a structured decision ledger and interactive visual dossier.
Free
Works with the AI Agents you already use
paper trail
Example session with this skill installed
I've been bouncing between tools for this app idea. Here are my ChatGPT and Claude exports. Merge them into a ledger and make an interactive dossier for my co-founder.
- Read your context and instructions
- Compiled the paper trail
- Generated the document
I have processed your chat exports and initialized the decision ledger. I identified 14 distinct decisions, 4 cost estimates, and 3 open questions regarding the database architecture and subscription tiers.
Project Dossier Compilation
I have normalized the transcripts from both sources and resolved three conflicting decisions regarding the frontend framework choice. The final dossier is ready for review.
| Metric | Detail |
|---|---|
| Sources Processed | 2 (ChatGPT-Export-2024-05.md, Claude-Chat-4.txt) |
| Total Events | 22 recorded in ledger.jsonl |
| Current Stage | Pre-MVP / Research |
| Key Highlights | SQLite selected for V1; $50/mo infra cap; Stripe integration accepted |
| Open Questions | 3 (Offline-sync effort, GDPR compliance, Logo designer) |
The interactive dossier has been generated. If you are in a filesystem-enabled environment, you can open the file below.
# build_dossier.py output
# JSON validation: PASSED
# HTML generation: SUCCESS
# File saved to: .dossier/dossier.html
Next steps
- Open
dossier.htmlin your browser to view the decision river and interactive timeline. - Review the "Open Questions" section; resolving these will remove the
assumptiontags from your cost estimates. - Use the command
checkpointafter our next brainstorming session to update the ledger.
dossier.html
PDF · document
Example file from a real run - the skill writes it into your workspace.
Connects securely to your tools. The creator never sees your data.
About this skill
Exploration sessions for new products often result in fragmented chat logs and forgotten logic. Decisions made in early brainstorming sessions are easily lost as context windows shift or conversations move between different AI platforms. This skill maintains a persistent, append-only decision ledger that survives long-term iteration and multi-tool research.
What it does
- Continuous capture records decisions, estimates, and reasoning into a persistent ledger as they happen.
- Multi-source ingestion merges exported transcripts from ChatGPT, Claude, and Gemini into a unified history.
- Backfill recovery analyzes existing chat history to reconstruct lost context and decision points.
- Interactive synthesis compiles the ledger into a self-contained HTML dossier with visual timelines and cost breakdowns.
- Audit tracking preserves the "why" behind rejected ideas, ensuring discarded paths are not repeated.
How it works
- Capture mode runs in the background, logging durable events like tech choices or budget estimates to
.dossier/ledger.jsonl. - Ingest mode processes external chat exports or markdown notes to normalize fragmented research into the central ledger.
- Compile mode triggers a replay of the ledger to resolve current states and builds a
dossier.jsonstructured for a specific audience. - Build step runs a validation script to ensure every claim has a confidence tag before generating the final interactive
dossier.html.
Frameworks & tools
This skill uses Python 3.9 scripts for data normalization and ledger management. It generates standard JSON and standalone HTML files that require no external dependencies or hosting to view.
Why this beats prompting it yourself
Manual summaries often hallucinate certainties or omit the specific reasons why an alternative was rejected. This skill uses a formal state machine to track decisions, ensuring that the final report is grounded in a verifiable audit trail rather than the model's shifting memory.
Use cases
- Tracking technical architecture pivots and trade-offs during a product build.
- Merging research from separate ChatGPT and Claude sessions into one brief.
- Generating a shareable, honest project status report for investors or stakeholders.
- Preserving the reasoning for "parked" features to avoid redundant future debates.
Known limitations
The HTML dossier requires the ledger to have specific evidence for sections to appear. It will refuse to build reports if critical numbers lack confidence tags or basis of estimation.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Agents & Orchestration.
Requires Python 3.9 for automated ledger scripts. The HTML dossier build will fail if required confidence tags or reasoning for estimates are missing from the ledger.
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
Free to use
No account required to browse
Trust & safety
Security scan
Security scan passed 23 days ago
- Free with an Agensi account
- Software
Python
Node.js
Git