Plugin Claude Code
Spec-first AI development harness — Go/TS/Java/PHP/Rust multi-language engineering standards, OWASP security gates, falsification-gated coding pipelines, 23 skills.
10 yesterday A
tokens not measured
original MIT
Instructions file CodexOpenCode
Instructions for LuisFelipeMoro/Harness-devkit, covering claude-devkit — agent instructions and harness-specific notes.
10 yesterday A 586 tokens
original MIT
Instructions file
Instructions for LuisFelipeMoro/Harness-devkit: This repository is the claude-devkit. To avoid drift, the engineering standards are maintained in a single canonical place — the coding-pipeline plugin's CLAUDE.md (the same file scripts/install-global.sh ships to /.claude/devkit/CLAUDE.md). It is included below and…
10 yesterday A 104 tokens
original MIT
Plugin Claude Code
11-agent spec-first coding pipeline (Analyst → PM → Architect → ScrumMaster → Coder → QA → Reviewer → StressTester → Tuner → Verdict → DevOps).
10 yesterday A
tokens not measured
original MIT
analyst
05Agent
Analyst agent (Mary) — produces a Project Brief from the task description.
10 yesterday A 19 tokens
original MIT
architect
06Agent
Architect agent (Winston) — produces the delivery file (docs/deliveries/delivery-{slug}-{key}.md) from the PRD.
10 yesterday A 33 tokens
original MIT
Agent
Bug Investigator agent (Sam) — diagnoses root cause and writes the failing RED test; never fixes implementation.
10 yesterday A 25 tokens
original MIT
Agent
Coder overlay — Backend (Amelia · server tier). Load with agents/coder.md core.
10 yesterday A 23 tokens
original MIT
Agent
Coder overlay — Frontend / Client (Amelia · UI tier). Load with agents/coder.md core.
10 yesterday A 26 tokens
original MIT
coder
10Agent
Coder core agent (Amelia) — implements to a frozen Test Case spec, then writes and falsifies the tests (Spec→Implement→Test→Falsify).
10 yesterday A 36 tokens
original MIT
devops
11Agent
DevOps agent (Ops) — generates infrastructure-as-code files (Dockerfile, compose, CI/k8s) after PRODUCTION READY verdict.
10 yesterday A 32 tokens
original MIT
pm
12Agent
PM agent (John) — produces a PRD from the Project Brief.
10 yesterday A 17 tokens
original MIT
qa
13Agent
QA agent (Quinn) — audits Amelia's tests for spec-completeness and falsifiability, runs quality gates, gates the pipeline. Never authors primary tests.
10 yesterday C 36 tokens
original MIT
reviewer
14Agent
Code Reviewer agent — reviews implementation code, outputs review score and findings.
10 yesterday A 17 tokens
original MIT
rote-adapter
15Agent
Autonomous adapter creation agent. Takes a request like "connect to Stripe" and runs the full discovery, analysis, research, auth, scoping, creation, and verification pipeline using rote CLI commands.
10 yesterday C 44 tokens
original MIT
Agent
Archive analytics specialist for DuckDB queries. Use for cost analysis, token usage, latency metrics, and usage patterns across MCP sessions.
10 yesterday A 30 tokens
original MIT
rote-datadog
17Agent
Datadog API V2 Collection specialist for Datadog V2 REST API providing programmatic access to observability, monitorin.... Use for single-domain datadog tasks. For workflows combining datadog with other services, use the main rote skill instead.
10 yesterday A 59 tokens
original MIT
rote-github
18Agent
GitHub v3 REST API specialist for GitHub's v3 REST API provides programmatic access to GitHub platform resource.... Use for single-domain github tasks. For workflows combining github with other services, use the main rote skill instead.
10 yesterday A 51 tokens
original MIT
scrum-master
19Agent
Scrum Master agent (Bob) — generates story files from Epic Manifest rows and the delivery file.
10 yesterday A 24 tokens
original MIT
stress
20Agent
Stress Tester agent — evaluates production resilience of implementation code + test suite.
10 yesterday A 16 tokens
original MIT
tuner
21Agent
Tuner agent (Tyler) — applies targeted MINOR/NIT fixes from a TUNER REQUEST.
10 yesterday A 25 tokens
original MIT
verdict
22Agent
Verdict agent — aggregates story/review/stress scores and issues the final production gate.
10 yesterday A 21 tokens
original MIT
SessionStart
23Hook
Runs when a session starts, executing session-bootstrap.sh via bash. From LuisFelipeMoro/Harness-devkit.
10 yesterday A
tokens not measured
copy · 59% MIT
PreToolUse
24Hook
Runs before the agent uses a tool for Read, Bash, Grep, Glob, Bash, Write, Edit and MultiEdit tool calls, executing env-guard.sh, destructive-guard.sh and secret-write-guard.sh via bash (3 commands). From LuisFelipeMoro/Harness-devkit.
10 yesterday A
tokens not measured
original MIT