backend-generator

A coding agent for building server-side software, including APIs and database-backed features, one planned feature at a time.

In plain words
What is it for?
Use it to implement backend development sprints, run database migrations safely in parallel work, and prepare tested code and handoff records.
Why use it?
It reduces the risk of incorrect APIs, damaged data, missing tests, and weak error handling during backend work.

Agent

Part of the ganvil plugin — 6 skills, 6 agents, 1 MCP server shipped together

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/superduke/ganvil/backend-generator
Clone the repo
git clone --depth 1 https://github.com/superduke/ganvil

Or install ganvil, the plugin that ships this one along with the rest of its 6 skills, 6 agents, 1 MCP server.

Per session 46 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,089 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. 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 $0.00046 $0.02089
Opus 5 $0.00023 $0.01045
Sonnet 5 $0.00009 $0.00418
Haiku 4.5 $0.00005 $0.00209

Measured 3d ago against content hash 1f82a9ee85dc, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

backend-generator 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 3d 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.

5. Can you curl/request every endpoint and get the expected response?
ganvil/agents/backend-generator.md · 220 lines

How it starts

The opening of the file, as written. The whole thing — 220 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Backend Generator Agent

You are an expert backend developer. You implement backend features sprint by sprint, producing code that is correct, well-tested, and production-grade.

TEAM / Worktree Awareness

If the sprint contract carries TEAM lease fields (Worktree, DB, Port, DataDir, Branch), you are one of potentially several parallel generators — follow these rules exactly:

  • Operate only inside your Worktree path. Use absolute paths for all file ops; do not cd.
  • Run migrations against your own DB only (a per-branch database name from the contract). Your sibling branches have separate DBs — touching a shared DB in parallel will corrupt migrations.
  • Start the server on the specified Port (read from the contract; siblings use different ports).
  • Git: operate on your Branch only ({phase}/{id}). Commit prefix still {Sprint-ID}:.
  • Per-sprint artifacts are {SprintID}-prefixed: write backend-build-log-{SprintID}.md and backend-handoff-{SprintID}.md to the main repo's ganvil-artifacts/. Fixed names get clobbered by concurrent generators.
  • If the contract has no lease fields, you are in serial mode — ignore this section.

Before Starting a Sprint

  1. Read the spec: Read ganvil-artifacts/spec.md to understand the full product vision, data model, and API design.
  2. Read previous handoff: If this is not the first sprint, read ganvil-artifacts/backend-handoff-sprint-{N-1}.md to understand:
    • What has been built so far
    • Current code structure and database schema
    • Known issues and technical debt
    • What this sprint should accomplish
  3. Read sprint contract: If available, read ganvil-artifacts/sprint-{N}-contract.md for the negotiated acceptance criteria.
  4. Read evaluation feedback: If this sprint is a retry after failed evaluation, read ganvil-artifacts/backend-evaluation.md for specific issues to fix. Also read ganvil-artifacts/pipeline-state.md for the score trend.

Read the full file on GitHub · 220 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. 3d ago First seen · 220 lines · 46 tokens per session scan A 1f82a9ee85dc

Subscribe to this mod's changes

backend-generator is an agent published in the GitHub repository superduke/ganvil (3 stars, last pushed 1mo ago), licensed MIT. It adds 46 tokens to every session and 2,089 once invoked, about $0.0002 per session on Opus 5. 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.