Daily copilot. Accumulated memory across all conversations, full vertical context. Use as default conversation mode: ask questions, verify decisions, discuss trade-offs. Lightweight startup: context loaded on demand, not automatically. Triggers: "copilot", "chat", "update me", "what do we have", "context".
Code implementer. Receives a detailed spec with file paths and patterns, implements precisely. Use after research is complete and a plan exists. Triggers: "implement this", "code this spec", "make these changes".
Codebase and knowledge base researcher. Read-only investigation: find files, trace data flows, search patterns, gather context. Use PROACTIVELY before implementation to understand existing code. Triggers: "research this", "find where", "trace the flow", "what pattern does this use".
Code reviewer. Analyzes changes against 12 quality dimensions. Read-only, never modifies code. Use after implementation to verify quality before commit. Triggers: "review these changes", "check this code", "quality check".
Capture business knowledge from Slack into the Knowledge Base. Two modes: (1) Single thread: pass a Slack URL to capture one discussion (2) Sweep: scan entire channels for Q&A threads, launch parallel agents per channel, catalog all findings Use when the user says "capture this", "salva esse conhecimento", "captura…
Review code quality with all 18 dimensions. Analyzes diff, gathers deep context, checks Known Gotchas, presents findings grouped by severity. If a PR exists, posts inline comments on GitHub. Use when the user says "code review", "review this PR", "review PR", "review", "pre-review", "check quality", or before opening…
Commit current work with a clean, conventional message. Use when the user says "commit", "save this", "git commit", "make the commit". Do NOT use for push (that requires separate approval).
Create a pull request with comprehensive technical format and monitor CI pipeline. Use when the user says "create pr", "open pr", "send the pr", "make the pr".
Generate daily standup notes from git history, session memory, and Jira. Use when the user says "daily", "generate daily", "create daily", "standup", "stand-up", "what did I do today", "daily notes".
Hypothesis-driven debugging workflow that requires runtime evidence before making fixes. Spins up a local log server so browser code can POST structured logs to a file the AI reads directly. Use when the user says "debug", "debug mode", "help me debug", "find this bug", "why isn't this working", "what's wrong with"…
Generate deployment checklist for a feature before merge. Use after PR is approved, before merge, or when the user says "deploy checklist", "checklist deploy", "pre-deploy", "ready to merge?", "checklist de deploy", "posso mergear?", "pronto pra deploy?", "can I merge?", "deploy readiness". Do NOT use for CI pipeline…
Remove AI-generated code slop (verbose comments, unnecessary abstractions, dead code, inconsistent patterns) from the current branch. Use after implementation and before commit/PR, or when the user says "deslop", "clean up AI code", "remove slop", "clean the code", "clean this up", "too much AI noise". Do NOT use for…
PlantUML diagram standard. Use when creating diagrams, flowcharts, sequence diagrams, entity diagrams, or when the user mentions "diagram", "diagrama", "plantuml", "sequence", "flowchart", "entity diagram", "C4".
Full feature development workflow with codebase discovery, knowledge base search, implementation plan, and guided coding. Use when the user says "feature", "new feature", "start development", "plan feature", "feature-dev", "implement", "build this feature", or pastes a Jira ticket description for implementation. Do…
Deep forensic investigation of bugs, incidents, or reported problems. Sweeps all sources (Jira, Slack, GitHub, Confluence, Codebase, Datadog/NR), cross-references data, reconstructs timeline, identifies root cause, and produces Investigation Case with Fix Plan. Use when the user says "investigation", "investigate"…
Capture a learning (bug, workaround, pattern, gotcha) and save to persistent memory. Use when the user says "learn", "save this", "remember this", "note this", "record this", "never forget this", or when a non-obvious bug fix, workaround, or integration pattern is discovered during development.
Validate feature with real test scenarios before commit. Verifies environment (Docker, VPN, DB), seeds test data, runs unit + integration tests, and executes manual smoke tests against local or staging. Use after /code-review and before /commit. Use when the user says "test", "testar", "test scenarios", "smoke test"…
Activate a self-critical, collaborative thinking mode. Use when the user says "thinking partner", "think together", "challenge me", "let's think", "think with me", or when the conversation needs deeper analysis rather than quick answers.
Validate infrastructure configuration consistency (env vars, Pulumi, config, schema). Use when adding env vars, changing Pulumi config, before deploy, or when the user says "validate infra", "check config", "valida infra", "env vars ok?", "check pulumi".
Validate database migration safety before merge. Checks rollback plan, backwards compatibility, performance impact, and lock duration. Use when creating migrations, altering tables, adding indexes, or when the user says "validate migration", "check migration", "migration safe?", "migration ok?", "can I run this…