Skip to content

Sathvikar01/Agent-engineering-skills / agent-development-workflow

agent-development-workflow

Verified

Coordinate the complete engineering lifecycle for substantial AI-agent or model-assisted workflow builds. Use when building, productionizing or substantially redesigning an agent, adding major tools/state/RAG/orchestration, or evaluating and hardening a whole agent system. Do not use for one bug, tiny prompt/schema/API changes, ordinary app work, simple deterministic utilities, conceptual agent explanations, or a narrow task owned by one specialist.

FileKindSize
SKILL.mdskill md7.7 KB
evals/evals.jsonconfig4.3 KB
evals/trigger-evals.jsonconfig3.4 KB
references/decision-and-defence.mdmarkdown1.3 KB

evals/trigger-evals.json

View on GitHub
[
  {
    "query": "Our receipt assistant works on one image. Help turn it into a reliable system that interprets messy receipts, validates reimbursements and emits supported explanations; build and test the whole path.",
    "should_trigger": true
  },
  {
    "query": "build me a support agent that can find orders and make authorized changes. I want the architecture, working implementation and evidence it behaves safely.",
    "should_trigger": true
  },
  {
    "query": "We have a chatbot prototype; productionize it with real state, tools and restart behavior rather than just a better prompt.",
    "should_trigger": true
  },
  {
    "query": "Substantially redesign our document agent: it loses evidence and loops. Audit architecture, implement a small corrected path, run failures and compare it to the old system.",
    "should_trigger": true
  },
  {
    "query": "Add retrieval, durable state and tool execution to the current one-call assistant. Work through the full engineering lifecycle and verify it.",
    "should_trigger": true
  },
  {
    "query": "I need an end-to-end hardening review of our agent, from model uncertainty and permissions to recovery, evals, traces and release evidence.",
    "should_trigger": true
  },
  {
    "query": "Create a semi-autonomous workflow for interpreting requests and preparing approvals. Need a measured baseline, vertical slice, implementation and failure tests.",
    "should_trigger": true
  },
  {
    "query": "Our autonomous research prototype is becoming a product. Plan and carry out its engineering upgrade, then explain why the final design survived evals.",
    "should_trigger": true
  },
  {
    "query": "The claims assistant needs a major capability addition: new evidence tools, persisted tasks and bounded routing. Coordinate design, implementation, execution and review.",
    "should_trigger": true
  },
  {
    "query": "Can you architect and build a complete model-assisted pipeline for messy invoices? Code controls money and approval; show actual verification before accepting it.",
    "should_trigger": true
  },
  {
    "query": "Fix the misspelling insufficent_evidence in our agent prompt.",
    "should_trigger": false
  },
  {
    "query": "The agent parser raises ValueError for an empty integer argument; fix that one bug and run the parser test.",
    "should_trigger": false
  },
  {
    "query": "Add evidence_version as an optional field to this existing proposal schema; leave other contracts alone.",
    "should_trigger": false
  },
  {
    "query": "Write one REST endpoint to list saved orders for our browser UI.",
    "should_trigger": false
  },
  {
    "query": "What is an AI agent? Explain it briefly to a beginner.",
    "should_trigger": false
  },
  {
    "query": "Make a deterministic cents-to-currency utility and its unit tests.",
    "should_trigger": false
  },
  {
    "query": "Tune just the search tool description so the LLM stops confusing it with delete_order.",
    "should_trigger": false
  },
  {
    "query": "Only compare latency and cost for these two existing model routes using the supplied numbers.",
    "should_trigger": false
  },
  {
    "query": "We already have our runtime; specify an exact email approval preview and expiry binding, no broader redesign.",
    "should_trigger": false
  },
  {
    "query": "Design the evidence ledger reconciliation for our existing answer service; retrieval, controller and release pipeline are fixed.",
    "should_trigger": false
  }
]
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 469kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 321kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 321kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 321kUpdated Claude CodeCodexVerifiedInstall