TudeOrangBiasa/opencode-workflow - 第 2 页

github-ops

Use when gitHub operations — issue triage, PR management, CI/CD debugging, release management, and security monitoring via gh CLI. Use only when managing GitHub issues, PRs, CI status, releases, contributors, stale items, or security alerts.

原文语言：英语

workflow-audit

Use when audit the opencode-workflow state — read opencode.json, verify symlinks, check repo sync, analyze recent sessions, surface mismatches. Use when user says audit, workflow status, check changes, verify sync, what did I change, opencode config status,…

原文语言：英语

handoff

Compact the conversation into a handoff document for another agent. Use when user says handoff, hand off, transfer context, pass to another agent, context compaction.

原文语言：英语

improve-codebase-architecture

Find deepening opportunities in a codebase, informed by the domain language in CONTEXT.md and the decisions in docs/adr/. Use when the user wants to improve architecture, find refactoring opportunities, consolidate tightly-coupled modules, or make a codebase…

原文语言：英语

to-spec

Turn the current conversation into a spec and publish it to the project issue tracker — no interview, just synthesis of what you've already discussed. Use when the user wants to create a spec or PRD from current context.

原文语言：英语

triage

Move issues and external PRs through a state machine of triage roles — categorise, verify, grill if needed, and write agent-ready briefs. Use when the user wants to triage an issue or PR, review incoming bugs or feature requests, or prepare issues for an AFK…

原文语言：英语

wayfinder

Plan a huge chunk of work — more than one agent session can hold — as a shared map of investigation tickets on your issue tracker, and resolve them one at a time until the way to the destination is clear.

原文语言：英语

write-a-skill

Create, audit, and write great agent skills — structure, progressive disclosure, bundled resources, and the principles that make a skill predictable. Use when user wants to create, write, build, or improve a skill, or asks about skill structure or quality.

原文语言：英语

ai-regression-testing

Use when regression testing for AI-assisted development — sandbox-mode API testing without DB, automated bug-check workflows, and patterns for AI blind spots where same model writes and reviews code. Use only when AI agents have modified API routes, after bug…

原文语言：英语

click-path-audit

Use when trace every user-facing button through its full state change sequence to find bugs where functions individually work but cancel each other out. Use only when systematic debugging found no bugs but users report broken UI, or after refactors touching…

原文语言：英语

diagnosing-bugs

Diagnosis loop for hard bugs and performance regressions. Use when the user says "diagnose"/"debug this", or reports something broken/throwing/failing/slow.

原文语言：英语

error-handling

Use when error-handling review checklist for typed errors, retry boundaries, user-safe messages, logging, API error envelopes, and swallowed exceptions. Use only when error paths are the task focus or code is at a high-risk boundary.

原文语言：英语

production-audit

Use when local-evidence production readiness audit — security, data integrity, operations, payments, UX. Score + blockers + next action. Use only for explicit launch/ship/prod readiness questions, not ordinary review.

原文语言：英语

tdd

Test-driven development. Use when the user wants to build features or fix bugs test-first, mentions "red-green-refactor", or wants integration tests.

原文语言：英语

team-handoff-quality

Team-ready handoff checklist for changes, verification evidence, version bump decisions, changelog entries, migration notes, rollback notes, and next-owner actions. Use when preparing work for teammates, AFK handoff, release notes, changelog updates, version…

原文语言：英语

defi-amm-security

Use when security checklist for Solidity AMM contracts — reentrancy, CEI ordering, donation/inflation attacks, oracle manipulation, slippage, admin controls, and safe reserve math. Use only when writing or auditing AMM, LP vault, or swap contracts.

原文语言：英语

evm-token-decimals

Use when prevent silent decimal mismatch bugs across EVM chains — runtime decimal lookup, chain-aware caching, bridged-token precision drift, and safe normalization. Use only when reading ERC-20 balances, calculating fiat values, or comparing amounts across…

原文语言：英语

nodejs-keccak256

Use only when computing Ethereum hashes in JavaScript/TypeScript — Node's sha3-256 is NIST SHA3, not Ethereum Keccak-256, and silently breaks selectors, signatures, and storage slots.

原文语言：英语

prediction-market-risk-review

Use when review prediction-market, basket, oracle, and trading-agent workflows for compliance, safety, data quality, privacy, and execution risk. Use only before any workflow handles venue auth, user portfolio data, API keys, or trade planning.

原文语言：英语

security-bounty-hunter

Use only when auditing web applications for security bounties — recon, XSS/CSRF/SQLi/SSRF testing, authentication bypasses, and report writing. Not for production security hardening.

原文语言：英语

security-review

Use when security review checklist for diffs touching auth, authorization, input handling, secrets, uploads, payments, PII, dependencies, CORS, or unsafe HTML. Use only when security-sensitive code or configuration changed.

原文语言：英语

openviking

Use when persist agent memory across sessions — store, retrieve, find, or update long-term context. Use when user says memory, remember, openviking, store, retrieve, save context, find memory, search memory. Persistent memory for AI agents with namespace…

原文语言：英语