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.
npx agentmods add agents/hundia/autospec/research-gsdgit clone --depth 1 https://github.com/Hundia/autospecWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/hundia/autospec/research-gsd)<a href="https://agentmods.dev/agents/hundia/autospec/research-gsd"><img src="https://agentmods.dev/badge/agents/hundia/autospec/research-gsd.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00000 | $0.05463 |
| Opus 5 | $0.00000 | $0.02731 |
| Sonnet 5 | $0.00000 | $0.01093 |
| Haiku 4.5 | $0.00000 | $0.00546 |
Grade A, and why
research-gsd scanned grade A with 1 finding 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 5d 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
<verify>curl -X POST localhost:3000/api/auth/login returns 200 + Set-Cookie</verify> How it starts
The opening of the file, as written. The whole thing — 326 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Research Report: GSD (Get Shit Done)
Researcher: A Date: 2026-03-20
What Is GSD
GSD — "Get Shit Done" — is a meta-prompting, context engineering, and spec-driven development system. It transforms AI coding assistants (Claude Code, Gemini CLI, Codex, OpenCode, Copilot, Antigravity) into reliable, structured development pipelines by managing context windows, orchestrating specialized sub-agents, and persisting project state to disk.
There are two generations of the project:
GSD v1 / get-shit-done-cc (npm: get-shit-done-cc, latest 1.27.0 as of 2026-03-20, 177 published versions, actively maintained — last publish was 5 hours before research time). This is the original "prompt-framework" variant: ~50 Markdown files installed as slash commands into ~/.claude/commands/gsd/ (or equivalent runtime config dirs). The LLM reads those Markdown instruction files and executes a structured workflow. No proprietary runtime — entirely built on top of existing AI assistants.
GSD v2 / gsd-2 (npm: gsd-pi, requires Node ≥ 22). This is a full TypeScript CLI harness built on the Pi SDK. It controls the LLM session directly: clears context between tasks, injects exactly the right files at dispatch, manages git branches, tracks token cost, detects stuck loops, recovers from crashes, and auto-advances through entire milestones without human intervention.
Both share the same underlying SDD philosophy: discuss → plan → execute → verify, with persistent state in structured Markdown files. This report covers both generations, noting where they differ.
Scale: 23k+ GitHub stars as of March 2026. Arguably the most widely adopted Spec-Driven Development tool in the AI-coding ecosystem.
LLM Integration Architecture
GSD v1 (get-shit-done-cc) — Pure Prompt Framework
GSD v1 does not call LLMs directly. Instead, it installs Markdown command files into the config directories of AI runtimes the user already has installed. The runtime's own LLM integration does the actual calling.
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.
- 5d ago First seen · 326 lines · 0 tokens per session scan A 3615e948dd93
research-gsd is an agent published in the GitHub repository Hundia/autospec (4 stars, last pushed 2mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 5,463 tokens. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
analyzer
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.
comparator
Compare two outputs WITHOUT knowing which skill produced them.
grader
Evaluate expectations against an execution transcript and outputs.
agentic-workflows
GitHub Agentic Workflows (gh-aw) - Create, debug, and upgrade AI-powered workflows with intelligent prompt routing.