ccevans/bobbycode

A full SDLC workflow for a solo developer — one person, a whole team of Claude Code agents.

This repository also configures its own agents. See what bobbycode tells them →

6Stars on the repository
84Mods indexed here, across every type
yesterdayLast push, which is what freshness is scored on
MITLicence, which decides whether bodies are shown

bobby-arch

01

ccevans/bobbycode

Agent Claude Code

Architecture discovery agent. Learns and documents the codebase so Bobby agents can implement tickets faster.

not rated 6 changed today A 24 tokens original MIT

bobby-build

02

ccevans/bobbycode

Agent Claude Code

Builds tickets using TDD. Commits code to the current branch.

not rated 6 changed today C 19 tokens original MIT

bobby-debug

03

ccevans/bobbycode

Agent Claude Code

Systematic root-cause debugging. Traces data flow, tests hypotheses, fixes with evidence.

not rated 6 changed today A 23 tokens original MIT

ccevans/bobbycode

Agent Claude Code

Product definition — generates the build blueprint and walks the human through it. Final job in the define workflow.

not rated 6 changed today A 28 tokens original MIT

bobby-define-brief

05

ccevans/bobbycode

Agent Claude Code

Product definition — interviews the founder and writes the brief. First job in the define workflow.

not rated 6 changed today A 25 tokens original MIT

ccevans/bobbycode

Agent Claude Code

Product definition — derives the feature map and locks the definition. Fourth job in the define workflow.

not rated 6 changed today A 26 tokens original MIT

ccevans/bobbycode

Agent Claude Code

Product definition — maps the primary journeys step by step. Third job in the define workflow.

not rated 6 changed today A 26 tokens original MIT

ccevans/bobbycode

Agent Claude Code

Product definition — derives personas from the brief. Second job in the define workflow.

not rated 6 changed today A 23 tokens original MIT

ccevans/bobbycode

Agent Claude Code

Design analysis — tears each reference down into extracted values. Second job in the design workflow.

not rated 6 changed today A 24 tokens original MIT

bobby-design-build

10

ccevans/bobbycode

Agent Claude Code

Design implementation — builds the real thing from the locked spec. Fifth job in the design workflow.

not rated 6 changed today A 24 tokens original MIT

bobby-design-check

11

ccevans/bobbycode

Agent Claude Code

Design check — independent live review against the spec and the slop checklist. Final job in the design workflow.

not rated 6 changed today A 27 tokens original MIT

bobby-design-mockup

12

ccevans/bobbycode

Agent Claude Code

Design mockups — builds options in each reference's system so the user can choose. Third job in the design workflow.

not rated 6 changed today A 31 tokens original MIT

ccevans/bobbycode

Agent Claude Code

Design research — gathers and cites the inspiration a design will be built from. First job in the design workflow.

not rated 6 changed today A 28 tokens original MIT

bobby-design-spec

14

ccevans/bobbycode

Agent Claude Code

Design spec — locks the agreed decisions into a versioned contract. Fourth job in the design workflow.

not rated 6 changed today A 25 tokens original MIT

bobby-docs

15

ccevans/bobbycode

Agent Claude Code

Updates project documentation after shipping. Keeps README, CLAUDE.md, and docs in sync with code.

not rated 6 changed today A 26 tokens original MIT

bobby-freewill

16

ccevans/bobbycode

Agent Claude Code

Takes a ticket from start to shipping alone, on deliberately few instructions. Runs on Fable by default.

not rated 6 changed today A 27 tokens original MIT

bobby-lighthouse

17

ccevans/bobbycode

Agent Claude Code

Lighthouse template audit. Sweeps Performance, Accessibility, Best Practices and SEO across page templates and proposes tickets on real gaps.

not rated 6 changed today A 30 tokens original MIT

bobby-performance

18

ccevans/bobbycode

Agent Claude Code

Performance benchmarking. Measures page load times, resource sizes, and Core Web Vitals.

not rated 6 changed today A 21 tokens original MIT

bobby-plan

19

ccevans/bobbycode

Agent Claude Code

Plans tickets — epic breakdown or refinement (plan.md + test-cases.md).

not rated 6 changed today A 20 tokens original MIT

bobby-pm

20

ccevans/bobbycode

Agent Claude Code

Product manager. Reviews the live application through the browser, identifies UX gaps and feature opportunities, shapes ideas into actionable tickets.

not rated 6 changed today A 29 tokens original MIT

bobby-qe

21

ccevans/bobbycode

Agent Claude Code

QE tester. Tests the live application through browser automation and API calls — never reads source code.

not rated 6 changed today A 24 tokens original MIT

bobby-review

22

ccevans/bobbycode

Agent Claude Code

Deep code review. Traces data flow, checks callers, catches regressions and LLM-typical mistakes.

not rated 6 changed today A 27 tokens original MIT

bobby-security

23

ccevans/bobbycode

Agent Claude Code

Security audit. OWASP Top 10 + STRIDE threat modeling on changed code.

not rated 6 changed today A 21 tokens original MIT

bobby-ship

24

ccevans/bobbycode

Agent Claude Code

Creates PR from current branch, waits for CI, merges, and deploys using whatever mechanism the repo needs. Ships all ready tickets.

not rated 6 changed today A 32 tokens original MIT

At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: