flowai-plugins

Use when the user asks to tailor the installed flowai skills, agents, hooks or AGENTS.md artifacts to this project's specifics. Not for first-time setup, template reconciliation, or edits to the project's own code.

Run prompts in Claude Code, OpenCode, Cursor, or Codex CLIs from this session — one IDE, fan-out, or model comparison. Relay the other runtime's stdout verbatim. Use on run in <ide>, compare <ide> vs <ide>, try on <model>, run across models.

Estimate total token usage and cost across conversation history, system prompts, and active rules. Use when the user asks how much context or how many tokens are in use, to audit context size, or to estimate session cost.

Use when the user asks a question to be answered from the project's memex (long-term knowledge bank for AI agents) rather than training data or the live web. Reads the index, follows one SALP-REF hop, files a cited answer back. Not for source code.

Use when the user asks to audit a memex (long-term knowledge bank for AI agents) for orphans, dead SALP REFs, missing sections, contradictions or index drift. Runs a structural check, layers LLM-judgement findings, auto-fixes trivia with `--fix`.

Automates browser interactions for web testing, form filling, screenshots and data extraction. Use when the user needs to navigate sites, interact with pages, fill forms, take screenshots, test web apps, or extract information from pages.

Use when the user asks whether the goals of the CURRENT session are met — lists everything they asked for, gathers evidence for each item and reports met or not met, changing nothing. Not for auditing how the session went — use reflect.

Use when a manager session is asked for the state of its child sessions — builds one board from the children's task files, branches and the host's session list, with the operator's next steps. Not for one session's own report (session-status).

Run Deno with the right flags — run/test permissions, `deno task` over raw commands, unstable opt-ins, deps via `deno add`/`outdated`/`remove`. Use when wiring or running Deno locally. Not for Deno Deploy or cloud hosting — use deploy.

Use when a manager session is asked to close its child sessions — checks that no child work is lost, moves follow-ups into the manager's task file, settles branches and worktrees, archives sessions. Not for finishing this session (finish-session).

Use when the user asks to commit the current changes — groups the diff into atomic conventional commits and updates the docs each change affects. Not for reviewing a diff, and not when the user asks for a review before the commit.

Configure and maintain Deno development commands (check, test, dev, prod). Use when the user wants to set up or update the standard command interface in deno.json and scripts/ directory.

Multi-directional deep research with sub-agents, scoring, source verification and cited synthesis. Use for comprehensive research, deep analysis, or evidence-backed answers. Produces an executive summary plus a markdown report.

Delegate a task to another AI IDE's CLI (codex / claude / opencode / cursor-agent) through an isolated-context subagent. Use when the user says delegate to <ide>, have <ide> do <task>, or offload to <ide>. For one-shot relay use `ai-ide-runner`.

Manage Deno Deploy cloud services using both `deno deploy` and `deployctl`. Use this skill for deploying projects, managing Build/Dev/Prod environments, monitoring logs, and troubleshooting issues like private npm dependencies.

Use when a judged benchmark or acceptance scenario fails and you need the cause before editing. Reads the judge's rendering, the raw transcript, the primitive the agent saw and the scenario, then reports a classified diagnosis without fixing.

Draw and edit Mermaid diagrams in Markdown. Use when the user wants to visualize processes, flows or sequences, or asks for a diagram. Load it before writing any Mermaid block, even when you know the syntax — that is why it gets skipped.

Design and build complete AI-IDE plugins bundling skills, MCP tools and hooks with packaging, manifests, marketplaces and per-IDE validation. Use when the user wants to create, design or package a multi-component plugin. Not for a standalone skill.

Guide for authoring an IDE slash command - one Markdown file the user runs as /name in Cursor, Claude Code or OpenCode. Use when users want to create or update such a command. Not for model-discovered skills.

Create and configure event hooks/plugins that manage agent behavior — command filtering, auditing, automation, script-based blocking of actions. Use when you need to create or configure a hook. Works across IDEs (Cursor, Claude Code, OpenCode).