hh-health-ai/healthcare-equity/edgar-forensics

Claude Code and agent skill by hh-health-ai/healthcare-equity

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

This skill should be used when the user asks for a "forensic scan", "quality of earnings", "red flags in the filings", "is the revenue real", "accounting risk", "channel stuffing", "reserve release", "non-GAAP bridge", or wants a filings-based integrity check on a healthcare issuer before initiating or shorting.

Install

npx vskill@latest install hh-health-ai/healthcare-equity/edgar-forensics

SKILL.md

EDGAR forensic scan

Run a healthcare-tuned accounting red-flag screen from XBRL company facts and filing text, and convert anything found into a testable claim rather than an insinuation.

Workflow

  1. Resolve the issuer to a CIK and pull company facts

(scripts/edgar_client.py --cik 0000320193 --facts).

  1. Run the ratio panel (scripts/forensic_ratios.py) across at least 12 quarters:

DSO, DIO, receivables growth versus revenue growth, gross margin trend, accruals ratio, cash conversion (CFO ÷ net income), capitalised software as a share of R&D, deferred revenue movement, and — where the taxonomy exposes it — rebate and chargeback accruals as a share of gross revenue.

  1. Apply the sector overlay. Match the issuer to a business model and read the

ratios through the right lens. The same DSO increase means something different for a hospital operator (payer mix or denial rates), a medtech (distributor loading), and a specialty pharma (channel and gross-to-net).

  1. Read the words, not just the numbers. Full-text search the filings

(--search) for: change in accounting estimate, restatement, material weakness, going concern, subsequent event, auditor change, segment redefinition, and any change in the wording of the revenue-recognition policy. A quietly reworded policy note is worth more than any ratio.

  1. Check the non-GAAP bridge. List every add-back for the last eight quarters and

flag any that recurs in five or more. Recurring "one-time" items are the most reliable single indicator that management is managing the number.

  1. Corroborate externally, or stand down. A revenue-recognition suspicion should be

tested against an independent volume series (rx-utilization → sdud-trx-proxy or provider-economics → hospital-margin-forensics). A red flag with no external corroboration is a question for the next call, not a thesis.

  1. State the falsifier. Write the specific disclosure that would clear the flag,

and the date it is next expected. This is what makes the brief actionable.

  1. Emit the brief.

Discipline

Rank findings by whether they affect cash, the trajectory, or only presentation. Presentation issues are common and rarely move a stock alone; cash issues are rare and usually terminal. Do not present a presentation issue with the rhetoric of a cash issue.

Never assert fraud. Describe the accounting choice, its effect on reported numbers, and the disclosure that would resolve it.

Not-automatic

A red-flag ratio does not license a fraud claim or a short. It licenses a specific question, a corroboration attempt, and a named falsifier.

Reference: references/healthcare-red-flags.md. Contract: ../../references/evidence-brief.md.

Security

  • Security scanPassed, 90 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

7-day trend score
0

Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/hh-health-ai/healthcare-equity/edgar-forensics/badge)](https://verified-skill.com/skills/hh-health-ai/healthcare-equity/edgar-forensics)

Something wrong with this skill or its page? Tell us.