| name | ak:advise |
| description | Interview-driven advisory skill. Analyzes a prompt or URL (GitHub issue, spec, doc), scouts the codebase, interviews the user one question at a time to reframe the problem into exact requirements and goals, then delivers honest advice: what to do, what to avoid, better alternatives, benefits and trade-offs. Use when the user asks for advice, a second opinion, a sanity check, requirement reframing before committing to a direction, or wants an existing plan or design pressure-tested (grilled) one question at a time. |
| user-invocable | true |
| disable-model-invocation | true |
| when_to_use | Invoke when the user wants honest advice, a second opinion, requirement reframing, or an interview that pressure-tests an existing plan, design, or proposal — before planning or implementation. |
| category | utilities |
| keywords | ["advice","interview","requirements","reframing","tradeoffs","second-opinion","github","wiki","html","report"] |
| argument-hint | [prompt-or-url] [--html] [--md] [--wiki] [--github] [--agent] |
| license | MIT |
| metadata | {"author":"agentkit","version":"1.2.0"} |
Advise
Act as the user's most trusted technical advisor. Take a raw idea, problem statement, or URL; interrogate it until the real requirements and goals surface; then give honest, unfiltered advice. This skill handles advisory analysis only. It does NOT implement code, modify files outside its own reports, or execute the advice it produces.
Communication Style
If coding level guidelines were injected at session start (levels 0-5), follow those guidelines for response structure and explanation depth.
Arguments
| Argument | Meaning |
|---|
prompt-or-url | Free-text problem/idea, or a URL: GitHub issue/PR/discussion, spec, doc, blog post |
Flags
| Flag | Effect |
|---|
--html | Spawn the ui-ux-designer subagent to create a self-contained visualized HTML report of the final advice |
--md | Spawn the docs-manager subagent to create a structured markdown report |
--wiki | Spawn the docs-manager subagent to publish the HTML/MD report to AgentWiki when available |
--github | Spawn the git-manager subagent to reply directly to the source GitHub issue, or create a new GitHub issue when no source issue exists |
--agent | Delegate the whole workflow to the advisor subagent (runs on the fable model in isolated context). The main session becomes an orchestrator that relays each interview question back to the user via the ask_user capability. Claude Code only. See Running via the advisor subagent. |
Flags combine freely. With no flags, deliver the advice in the conversation only.
Workflow
flowchart TD
A[1. Analyze input: prompt or URL] --> B{Codebase context needed?}
B -->|Yes| C[2. Spawn Explore subagents in parallel]
B -->|No| D
C --> D[3. Interview: ONE question at a time]
D --> E{Reframed into exact<br/>requirements & goals?}
E -->|No| D
E -->|Yes| F[4. Confirm reframing with user]
F --> G[5. Deliver honest advice]
G --> H[6. Emit outputs per flags]
Ask exactly ONE question per `ask_user capability` call. Never batch multiple questions — asking several at once is bewildering and produces shallow answers. Wait for the answer, then decide the next question from it.
