corvid-agent

Git workflows — branching, committing, worktrees, branch isolation, merge strategies. Trigger keywords: git, commit, branch, merge, rebase, worktree, pull, push, cherry-pick.

Testing — writing and running tests with bun:test, test patterns, validation. Trigger keywords: test, testing, unit test, integration test, bun test, spec, assertion.

Pre-commit verification — tsc type checking, test suite, spec invariant checks. Trigger keywords: verify, validate, check, tsc, typecheck, spec check, pre-commit.

Discord messaging — how to send and receive messages, bridge architecture, user identity, formatting. Trigger keywords: discord, discord message, discord reply, discord user, discord bridge, send discord.

Cross-channel messaging — routing rules, channel affinity, how to send messages on each channel, safety rules. Trigger keywords: message, send message, reply, respond, channel, routing, corvid_send_message.

Use this skill when you need to discover remote agents, fetch their capabilities via A2A Agent Cards, or invoke tasks on remote A2A-compatible agents. Triggers include "discover agent", "find agent", "agent card", "invoke remote agent", "send task to agent", "A2A", or any reference to inter-agent communication beyond simple messaging.

Use this skill when the user wants to find code symbols (functions, classes, types), trace references across a codebase, or navigate code structure. Triggers include "find function", "where is this defined", "find references", "symbol search", "code navigation", "what calls this", or any request to locate or trace code entities.

Use this skill when you need to read, write, edit, or search files, run shell commands, or list directory contents. These are the core file operation and command execution tools. Triggers include any coding task, file manipulation, running tests, building, or executing commands.

Use this skill when you need to look up a contact by name or platform ID to resolve cross-platform identities. Triggers include "who is", "find contact", "look up user", "resolve identity", "Discord ID", "GitHub username", or any request to identify someone across platforms.

Use this skill when the user wants to check credit balances, grant credits to a wallet, or configure the credit system. Triggers include "check credits", "credit balance", "grant credits", "free credits", "credit config", "billing", or any reference to the CorvidAgent credit system.

Database operations — SQLite migrations, bun:sqlite queries, schema patterns. Trigger keywords: database, sqlite, migration, schema, query, table, db.

Use this skill when the user wants to check codebase health metrics, view health trends over time, or monitor improvement cycles. Triggers include "health check", "health trends", "code quality", "metrics", "improvement cycle", "codebase health", or any reference to tracking code quality over time.

Use this skill when the agent needs to notify the server owner, ask the owner a question, or configure notification channels (Discord, Telegram, GitHub, etc). Triggers include "notify owner", "ask owner", "alert the admin", "configure notifications", "set up Discord notifications", or any need to communicate with the agent's operator.

Use this skill when the user wants to list available projects, check which project is active, or understand the project context. Triggers include "list projects", "what projects", "current project", "which project", "project info", or any reference to project selection and context.

Use this skill when you need to manage the repository blocklist — blocking, unblocking, or checking if repos are blocked from agent operations. Triggers include "block repo", "unblock repo", "blocklist", "blocked repositories", or any reference to restricting agent access to specific repositories.

REST API development — adding endpoints, route patterns, middleware, request/response conventions. Trigger keywords: api, endpoint, route, REST, HTTP, handler, middleware, request, response.

Use this skill when the user wants to search the web, research a topic, or find current information online. Triggers include "search for", "look up", "find out about", "research", "what's the latest on", "web search", or any request requiring up-to-date information beyond your training data.

Swift development — writing, reviewing, and rating Swift code for iOS/macOS. Trigger keywords: swift, iOS, macOS, xcode, swiftui, uikit, swift code review.

Telegram messaging — bot bridge, voice notes, message handling, STT/TTS integration. Trigger keywords: telegram, telegram bot, telegram message, telegram voice.

Voice integration — TTS via OpenAI tts-1, STT via Whisper, audio caching. Trigger keywords: voice, tts, stt, text to speech, speech to text, whisper, audio.

Use this skill when the user wants to create an autonomous coding task, spawn an agent to implement a feature or fix a bug, or automate code changes with PR creation. Triggers include "create a work task", "implement this feature", "fix this bug autonomously", "spawn an agent to", "create a PR for", or any request to delegate coding work to an autonomous agent session.

Use this skill when the user wants to send or receive messages on AlgoChat, communicate with other AI agents, or use Algorand-based messaging. Triggers include "send a message", "message an agent", "chat with", "reply to", "list agents", "AlgoChat", or any reference to on-chain messaging between agents.

Use this skill when the user wants to register an agent in the directory, search for agents, discover agent capabilities, or manage agent registry entries. Triggers include "register agent", "find agents", "agent directory", "flock directory", "search for agents", "agent registry", "discover agents", "heartbeat".

Use this skill when the user wants to interact with GitHub through CorvidAgent — creating PRs, reviewing code, managing issues, starring repos, or forking repositories. Triggers include "create a PR", "review pull request", "create an issue", "star this repo", "fork", "list PRs", "code review", or any GitHub repository operation.

Use this skill when the user wants to save, recall, or search agent memories. Memories are encrypted and stored on the Algorand blockchain, persisting across sessions. Triggers include "remember this", "save a memory", "recall", "what did I save", "agent memory", "on-chain memory", "persistent storage".

Use this skill when the user wants to coordinate multiple agents, launch a council for decision-making, invoke remote agents, or create multi-step workflows. Triggers include "launch a council", "multi-agent", "invoke remote agent", "agent-to-agent", "A2A", "coordinate agents", "deliberation", "consensus", or any request involving multiple AI agents working together.

Use this skill when the user wants to check agent reputation, verify trust levels, publish attestations, or track codebase health trends. Triggers include "reputation", "trust level", "attestation", "verify agent", "health trends", "code quality score", or any reference to on-chain agent trust and verification.

Use this skill when the user wants to schedule recurring tasks, automate PR reviews, set up cron jobs for agents, create workflows, or automate any periodic agent activity. Triggers include "schedule", "cron", "every day", "weekly", "automate", "recurring", "workflow", "run daily", "set up automated reviews".

Use this skill when the user wants to build, deploy, test, or interact with Algorand smart contracts. Triggers include "deploy a contract", "write a smart contract", "call a method", "create an ASA", "create a token", "application state", "ABI", "opt in to app", "PuyaTs", "PuyaPy", "Algorand contract", "mainnet deploy", "testnet deploy", or any blockchain development task on Algorand.