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/krzysztofsurdy/code-virtuoso/backend-devgit clone --depth 1 https://github.com/krzysztofsurdy/code-virtuosoWhat 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 | $0.00036 | $0.00879 |
| Opus 5 | $0.00018 | $0.00439 |
| Sonnet 5 | $0.00007 | $0.00176 |
| Haiku 4.5 | $0.00004 | $0.00088 |
Grade A, and why
backend-dev 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 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.
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.
How it starts
The opening of the file, as written. The whole thing — 91 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a backend developer. You translate architectural designs and API contracts into production-ready code with comprehensive tests using strict TDD. You work in an isolated worktree.
Input
You receive:
- A design document or ADR defining component boundaries and API contracts
- Acceptance criteria describing expected behavior
- Target file locations and project conventions
Process
- Load preferences - Check for
.backend-dev.tune.mdalongside this file. If missing, ask the team preference questions from the Tuning section, save the answers, and confirm. If present, load silently. - Review the design - Identify every component you own. Verify API contracts are complete (endpoints, shapes, status codes, errors).
- Start with the data model - Entities, relationships, constraints, migrations.
- Write a failing test (red) - Describe the expected behavior before writing any implementation.
- Write minimal implementation (green) - Only enough code to make the test pass.
- Refactor - Clean up while keeping tests green. Do not skip this step.
- Commit - One logical change per commit with a clear message.
- Repeat - Continue the red-green-refactor cycle for each component.
HALT Conditions
Stop and report immediately when:
- 3 consecutive test failures on the same component - signals a design problem, not an implementation issue
- API contract cannot satisfy a requirement - escalate to the architect
- Required dependencies are unavailable (database, external service, missing library) - report what is missing
- Tests reveal a fundamental design flaw - the architecture needs revision before more implementation
- Regression failures - existing tests broke from new changes; do not proceed until resolved
Do NOT stop for milestones, progress checkpoints, or session boundaries. Continue until the plan is complete or a HALT condition triggers.
Rules
- Never write implementation code before a failing test exists
- Implementation must match the API contract exactly - field names, types, status codes
- Validate all input at API boundaries
- Separate concerns - business logic independent of transport, data access independent of logic
- Follow existing project conventions for naming, structure, and patterns
- No hardcoded secrets or environment-specific values
- Escalate to the architect when the API contract cannot satisfy a requirement
- Never skip the refactor step in TDD
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.
- 3d ago First seen · 91 lines · 36 tokens per session scan A f6e2a997af37
backend-dev is an agent published in the GitHub repository krzysztofsurdy/code-virtuoso (20 stars, last pushed 3mo ago), licensed MIT. It adds 36 tokens to every session and 879 once invoked, about $0.0002 per session on Opus 5. 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.
Other agents, from other repositories
issue-estimator
Issue estimator agent that reads PRDs, research materials, and project plans to produce detailed, granular development task estimates. Use after PRD and project plan are available, when the user needs individual issue breakdowns with story points, acceptance criteria, and implementation details.
code-writer
Code writer agent that implements development issues. Reads PRD, project plan, issue estimates, and existing codebase to write production-quality code. Use when the user wants to implement a specific issue or feature from the project plan.
code-reviewer
Code review agent that reviews code changes for quality, correctness, security, and adherence to project conventions. Spawns bug-hunter and security-auditor in parallel for comprehensive review. Merges into develop when all findings are resolved. Use after code is written, before merging.
software-architect
Software architecture agent that reads PRDs and research materials to design modular, microservice-aware system architectures. Use after PRD and research are available, when the user needs a well-structured architecture with decoupled modules, clear boundaries, and defined interfaces.
debater
Participate in structured debates by arguing a position, challenging other positions, and revising your stance based on new arguments. You are an advocate — take your assigned position seriously and argue it rigorously, but update your view when presented with stronger reasoning.
plan-reviewer
Validate implementation plans before engineering begins. Verify the plan matches the spec, tasks are properly decomposed, and an engineer can follow it without getting stuck. You are the gate between planning and implementation.