adhoc-edits-workflow

Applies one-off edits to assembled deal memos through a four-step research, section, assembly, and export pipeline.

Updated Aug 15, 2026
One-click install
npx skills add https://github.com/junkyardhummingbird/investment-memo-orchestrator --skill adhoc-edits-workflow-junkyardhummingbird
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: adhoc-edits-workflow
Source: https://github.com/junkyardhummingbird/investment-memo-orchestrator/tree/main/context-v/agent-skills/adhoc-edits-workflow
Command: npx skills add https://github.com/junkyardhummingbird/investment-memo-orchestrator --skill adhoc-edits-workflow-junkyardhummingbird

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? One-off changes to an already-assembled investment memo often get lost because edits made directly to the final draft are overwritten by the citation assembler, and citations added only in section files get stripped, leaving broken footnotes in the exported PDF. ## Core Features & Use Cases - Four-Step Edit Pipeline: Enforces a strict order of adding citation definitions to research files, integrating prose into section files, running the citation assembler, and exporting HTML then PDF. - Citation Integrity Rules: Requires semantic citation IDs, verification of facts via authoritative sources, and checks for missing definitions after assembly. - Use Case: When a user says "Add Paramount to the customer list" or "Fix the seed round — it was a SAFE bridge," this workflow ensures the change survives re-assembly and appears correctly in the final exported PDF. ## Quick Start Apply the ad-hoc edits workflow to add a new customer to the customer list section of the current deal memo and re-export the PDF.

Frequently Asked Questions about adhoc-edits-workflow

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

FAQPage Schema
How do I make a one-off edit to an existing investment memo?

Add the citation definition to the matching research file first, then integrate the prose into the correct section file using the same semantic citation ID. Run the citation assembler to rebuild the final draft, then export HTML and PDF with the branded export script.

Why did my memo edit disappear after re-assembly?

Edits made directly to the final draft file are overwritten because it is a build artifact regenerated by the assembler. Prose must live in the section files and citation definitions in the research files to survive re-assembly.

Why does my exported PDF show literal footnote text like [^id]?

A literal footnote marker means the inline citation reference has no matching definition anywhere, so Pandoc renders it as plain text. Add the citation definition to the appropriate research file and re-run the assembler before exporting.

Can I use numeric citation IDs in memo section files?

No, hand-typed numeric IDs are unsafe because the assembler renumbers all citations sequentially on every run. Always use semantic citation IDs so references remain stable across assembly passes.

When should I use this workflow instead of re-running the full memo pipeline?

Use it when the change is scoped to one or a few specific edits, such as adding a customer, correcting a financing detail, or inserting a chart. A full pipeline re-run is only warranted for broad regeneration of the memo.