LuisFelipeMoro

52 mods across 1 repository, 10 stars between them.

PostToolUse

25

LuisFelipeMoro/Harness-devkit

Hook

Runs after a tool call finishes for Bash, Bash, Write, Edit and MultiEdit tool calls, executing pr-review-responder.sh and session-tracker.sh via bash (2 commands). From LuisFelipeMoro/Harness-devkit.

10 2d ago A tokens not measured original MIT

Stop

26

LuisFelipeMoro/Harness-devkit

Hook

Runs when the agent finishes a response, executing delivery-gate.sh via bash. From LuisFelipeMoro/Harness-devkit.

10 2d ago A tokens not measured copy · 58% MIT

analysis

27

LuisFelipeMoro/Harness-devkit

Skill Claude CodeCodex

Use when you need to analyze requirements, assess code state, or evaluate context — runs Analyst then PM to produce product-brief.md and PRD.md without architecture or execution steps.

10 2d ago A 38 tokens original MIT

architecture

28

LuisFelipeMoro/Harness-devkit

Skill Claude CodeCodex

Use when designing architecture for a domain or feature outside the full multi-agent pipeline. Produces tech stack, security architecture, component design, ADRs, and Mermaid data-flow diagram.

10 2d ago A 38 tokens original MIT

bug-fix

29

LuisFelipeMoro/Harness-devkit

Skill Claude CodeCodex

Use when investigating and fixing a behavioral bug — dispatches Sam (Bug Investigator) and Amelia (Coder) as sub-agents to keep exploration out of main context. Trigger phrases — "bug", "fix", "broken", "not working", "wrong behavior", "unexpected", "crash", "regression", "debug", "fails with".

10 2d ago A 75 tokens original MIT

multi-agent

30

LuisFelipeMoro/Harness-devkit

Skill Claude CodeCodex

Use when running the full multi-agent coding pipeline for a large feature or epic — runs all 11 agents (Analyst, PM, Architect, ScrumMaster, Coder, QA, Reviewer, StressTester, Verdict, Tuner, DevOps) from planning through delivery. Trigger phrases — "build", "new feature", "epic", "implement from scratch"…

10 2d ago A 85 tokens original MIT

planning

31

LuisFelipeMoro/Harness-devkit

Skill Claude CodeCodex

Use when you need to break down analyzed requirements into an actionable execution plan — runs Architect to produce the delivery file and the Epic or Task Manifest. Coupled to the coding pipelines; produces plan artifacts only, never triggers implementation.

10 2d ago A 47 tokens original MIT

task

32

LuisFelipeMoro/Harness-devkit

Skill Claude CodeCodex

Use when given a single coding task to implement — skips planning, runs Architecture then decomposes into independently-testable sub-tasks, coding, QA with quality gates, review, stress testing, and verdict per sub-task.

10 2d ago A 46 tokens original MIT

devtools

33

LuisFelipeMoro/Harness-devkit

Plugin Claude Code

Developer tools — architectural health review, business/technical analysis, adversarial plan grilling, session handoff, skill authoring, rote adapter creation, rote flow automation.

10 2d ago A tokens not measured original MIT

business-analysis

34

LuisFelipeMoro/Harness-devkit

Skill Claude CodeCodex

Use when mapping business rules, constraints, and presentation logic of a domain. Produces user capability map, validation rules, presentation rules, and external integrations in Brazilian Portuguese.

10 2d ago A 37 tokens original MIT

grill-me

35

LuisFelipeMoro/Harness-devkit

Skill Claude CodeCodex

Use when stress-testing a plan, design, or architectural decision before committing to it. Asks one hard question at a time with a concrete recommendation. Stops when user says "done" or "enough". Trigger phrases: "grill me", "challenge this", "stress-test", "poke holes", "pick this apart", "interview me about".

10 2d ago A 77 tokens original MIT

handoff

36

LuisFelipeMoro/Harness-devkit

Skill Claude CodeCodex

Use when ending a long session or switching context. Compacts the conversation into a structured handoff document saved to /tmp. Trigger phrases: "handoff", "wrap up", "end session", "save context", "create handoff", "compact this session", "summarize for next session".

10 2d ago A 65 tokens original MIT

LuisFelipeMoro/Harness-devkit

Skill Claude CodeCodex

Use when deepening architectural quality of a codebase. Finds coupling, boundary violations, missed abstractions, and SRP failures. Produces HTML report in /tmp. Offers ADRs and grilling loop. Trigger phrases: "improve architecture", "architectural review", "find coupling", "refactor architecture", "codebase health"…

10 2d ago A 84 tokens original MIT

rote-adapter

38

LuisFelipeMoro/Harness-devkit

Skill Claude CodeCodex

Use when creating a NEW integration adapter from scratch — discovers the API, negotiates auth, scopes endpoints, builds, and verifies the adapter in 8 autonomous phases. Use /rote for running existing adapters/flows. Trigger phrases: "build adapter", "create integration", "new connector", "connect to X for the first…

10 2d ago A 87 tokens original MIT

rote

39

LuisFelipeMoro/Harness-devkit

Skill Claude CodeCodex

Use rote BEFORE calling any MCP server or CLI tool directly. rote wraps installed adapters (MCP servers and CLI-based tools) and adds flow reuse, response caching, and crystallized workflows. Trigger examples: "list my open tickets", "what should I work on next", "fetch issues from the project", "show calendar…

10 2d ago A 118 tokens original MIT

technical-analysis

40

LuisFelipeMoro/Harness-devkit

Skill Claude CodeCodex

Use when mapping technical contracts and infrastructure of a domain. Read-only — produces HTTP routes, interfaces, external integrations, and infrastructure overview without modifying code.

10 2d ago A 33 tokens original MIT

write-a-skill

41

LuisFelipeMoro/Harness-devkit

Skill Claude CodeCodex

Use when creating a new skill. Scaffolds proper structure with clear input/output/boundary per phase. Skill files must be concise — under 100 lines, overflow to references/. Trigger phrases: "write a skill", "create skill", "add skill", "new skill", "scaffold skill".

10 2d ago A 67 tokens original MIT

engineering

42

LuisFelipeMoro/Harness-devkit

Plugin Claude Code

Engineering quality skills — OWASP security review, quality gates, code-review-gate (gate + reviewer for any non-pipeline code), DB migration, observability, performance profiling, release management.

10 2d ago A tokens not measured original MIT

code-review-gate

43

LuisFelipeMoro/Harness-devkit

Skill Claude CodeCodex

Gate and review any code generated outside a full pipeline (inline spec-first sessions, direct edits, ad-hoc fixes) by clearing all quality gates and then routing to the Reviewer agent. Gates must pass before the Reviewer runs. Trigger phrases — "gate and review", "pre-push check", "ready to push", "sign off my code"…

10 2d ago A 89 tokens original MIT

database-migration

44

LuisFelipeMoro/Harness-devkit

Skill Claude CodeCodex

Write or review a database migration safely with additive-only forward changes, a mandatory reversible down migration, zero-downtime patterns for large tables, and lock-acquisition analysis. Trigger phrases — "migration", "schema change", "add column", "alter table", "add index", "backfill", "drop column", "rename…

10 2d ago A 72 tokens original MIT

observability

45

LuisFelipeMoro/Harness-devkit

Skill Claude CodeCodex

Add structured logging, metrics, or distributed tracing to a service, covering Go (zap + OpenTelemetry) and TypeScript (pino + OpenTelemetry) with enforced log-field standards and trace-context propagation. Trigger phrases — "logging", "metrics", "tracing", "observability", "instrument", "OpenTelemetry", "structured…

10 2d ago A 76 tokens original MIT

LuisFelipeMoro/Harness-devkit

Skill Claude CodeCodex

Investigate performance issues in Go services with pprof setup, profile capture, an analysis workflow, and common optimizations. Trigger phrases — "performance", "slow", "profiling", "optimize", "latency", "pprof", "benchmark", "bottleneck", "high CPU", "memory leak".

10 2d ago A 70 tokens original MIT

quality-gate

47

LuisFelipeMoro/Harness-devkit

Skill Claude CodeCodex

Verify code is shippable by detecting the project stack and reporting PASS or FAIL for every gate before handoff. Trigger phrases — "quality gate", "run gates", "CI check", "lint", "coverage", "run tests", "check before PR".

10 2d ago A 57 tokens original MIT

release-management

48

LuisFelipeMoro/Harness-devkit

Skill Claude CodeCodex

Cut a release by determining the semver bump from conventional commits, updating CHANGELOG.md, tagging, pushing, and drafting a GitHub release. Trigger phrases — "release", "cut a release", "bump version", "changelog", "tag", "semver", "publish", "release notes".

10 2d ago A 66 tokens original MIT