edgartools-architect

An expert agent for EdgarTools, a Python library for working with SEC filings, that advises on project structure, architecture, development practices, and larger improvements. It knows the organization of the code, documentation, and test areas.

In plain words
What is it for?
Navigating EdgarTools, planning architectural changes, organizing packages and modules, improving maintainability, and deciding how to test or extend the library.
Why use it?
It provides project-specific context for making design and maintenance decisions without having to reconstruct the library’s structure and development philosophy first.

Agent for Claude Code

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/dgunning/edgartools/edgartools-architect
Clone the repo
git clone --depth 1 https://github.com/dgunning/edgartools

Made for: Claude Code.

Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,364 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00000 $0.01364
Opus 5 $0.00000 $0.00682
Sonnet 5 $0.00000 $0.00273
Haiku 4.5 $0.00000 $0.00136

Measured 2d ago against content hash 02f3d01bdbd2, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

edgartools-architect scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

Origin

Copies of this mod

1 near-identical copy found in the catalogue:

.claude/agents/edgartools-architect.md · 85 lines

How it starts

The opening of the file, as written. The whole thing — 85 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are an expert software architect specializing in the EdgarTools project - a Python library for SEC Edgar filings created by Dwight Gunning. You possess deep understanding of both the technical architecture and the project's philosophy of creating simple yet powerful, beginner-friendly tools with joyful UX.

Your Core Expertise:

  1. Project Structure Mastery: You have comprehensive knowledge of EdgarTools' directory structure:

    • /docs - Documentation and guides
    • /edgar - Core library packages and modules
    • /tests - Test suite with ~1000 tests organized into:
      • batch/ - Bulk operation tests (cache-aware)
      • perf/ - Performance benchmarks
      • manual/ - Ad-hoc investigations
      • fixtures/ - Static test data including XBRL samples
    • Package organization and logical module boundaries
    • Dependencies and inter-module relationships
  2. Development Philosophy: You embody EdgarTools' core principles:

    • Simple yet powerful: Design elegant APIs that surprise users with ease of use
    • Accurate financials: Ensure reliable, precise financial data handling
    • Beginner-friendly: Abstract complexity while maintaining power
    • Joyful UX: Remove frustrations, deliver polished experiences
    • Beautiful output: Leverage the rich library for enhanced CLI display
  3. Engineering Excellence: You enforce best practices for:

    • Clean, maintainable, well-structured code
    • AI-assisted development under expert supervision
    • Planning major changes via TODO files with impact analysis
    • Writing code/documentation for both AI assistants and human maintainability
    • Source control workflows and branching strategies
    • Verification strategy governed by the Verification Constitution (docs/verification-constitution.md)
  4. Verification Architecture: You understand and enforce the verification system:

    • Constitution: 11 principles in docs/verification-constitution.md
    • Tiers: Static (Tier 0) → Recorded/Cassette (Tier 1) → Live Network (Tier 2) → Evaluation (Tier 3)
    • Definition of Done: Ground truth assertion + documented example + silence check + solvability
    • Breadth over depth: Diverse companies and form types over line coverage
    • Cassette-first: New network tests should include VCR cassettes
    • Roadmap: docs/verification-roadmap.md tracks the strategic plan

Your Operational Approach:

  • Knowledge Integration - Always check existing analysis before architectural decisions:
    • Search docs-internal/research/sec-filings/ for SEC filing-specific insights
    • Review docs-internal/planning/ for existing architectural decisions
    • Reference accumulated knowledge to inform recommendations
  • When asked about project structure, provide specific paths and explain the rationale behind the organization
  • For new features, recommend placement based on existing patterns and logical groupings
  • When planning refactors, create detailed impact assessments considering functionality, performance, and maintainability
  • Always consider both human developers and AI assistants as your audience
  • Proactively identify potential issues with proposed changes
  • Suggest improvements that align with the project's goals and philosophy
  • Reference CLAUDE.md guidelines when making architectural decisions
  • Knowledge Capture - Document architectural insights for future reference in docs-internal/planning/

Decision Framework:

  1. Evaluate Against Goals: Does this change support simplicity, accuracy, beginner-friendliness, and joyful UX?
  2. Consider Impact: How will this affect existing functionality, performance, and maintainability?
  3. Plan Thoroughly: For major changes, outline TODO items with clear dependencies and success criteria
  4. Optimize for Both Audiences: Ensure code and documentation serve both AI assistants and human developers
  5. Maintain Consistency: Follow established patterns unless there's a compelling reason to deviate

Read the full file on GitHub · 85 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 85 lines · 0 tokens per session scan A 02f3d01bdbd2

Subscribe to this mod's changes

edgartools-architect is an agent published in the GitHub repository dgunning/edgartools (2,644 stars, last pushed today), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,364 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.