counselors（counselors 搜索）

Counselors — Multi-Agent Review Skill

Fan out a prompt to multiple AI coding agents in parallel and synthesize their responses.

Arguments: $ARGUMENTS

If no arguments provided, ask the user what they want reviewed.

Phase 1: Context Gathering

Parse $ARGUMENTS to understand what the user wants reviewed. Then auto-gather relevant context:

1. Files mentioned in the prompt: Use Glob/Grep to find files referenced by name, class, function, or keyword
2. Recent changes: Run git diff HEAD and git diff --staged to capture recent work
3. Related code: Search for key terms from the prompt and read the most relevant files (up to 5 files, ~50KB total cap)

Be selective — don't dump the entire codebase. Pick the most relevant code sections.

Phase 1b: Context7 Staleness Scan

Detect the project's key technologies from package.json, CLAUDE.md, tsconfig.json, or config files (if not already clear from Phase 1 context). For up to 5 key libraries/frameworks:

1. mcp__plugin_compound-engineering_context7__resolve-library-id — get the Context7 library ID
2. mcp__plugin_compound-engineering_context7__query-docs — fetch 2-3 relevant snippets focused on APIs, configuration, and breaking changes

Limits:

• Cap total reference documentation at ~8,000 tokens. Trim the least relevant snippets if exceeded.
• If a library isn't found in Context7: add a line like [library name] (docs not verified) to the REFERENCE DOCUMENTATION block and continue.
• If no specific libraries are identifiable from context: skip this phase entirely.

Build a REFERENCE DOCUMENTATION block with library name + version per entry. This block will be included in the prompt (Phase 3) so reviewers can flag outdated patterns.

Phase 2: Agent Selection

Default agents: or-claude-opus, or-gemini-3.1-pro, or-codex-5.4

1. Use defaults unless the user overrides. If $ARGUMENTS does not contain agent-selection instructions (e.g. "use all agents", "add codex", "only gemini"), skip directly to the confirmation step with the defaults.
2. If the user requests different agents (in $ARGUMENTS or via follow-up), discover available agents by running via Bash: counselors ls Print the full output, then ask the user to pick using AskUserQuestion. If 4 or fewer agents: Use AskUserQuestion with multiSelect: true, one option per agent. If more than 4 agents: AskUserQuestion only supports 4 options. Use these fixed options: Do NOT combine agents into preset groups (e.g. "claude + codex + gemini"). Each option must be a single agent or "All".

- Option 1: "All [N] agents" — sends to every configured agent - Option 2-4: The first 3 individual agents by ID - The user can always select "Other" to type a comma-separated list of agent IDs from the printed list above

1. MANDATORY: Confirm the selection before continuing. Echo back the exact list you will dispatch to: Dispatching to: or-claude-opus, or-gemini-3.1-pro, or-codex-5.4 Then ask the user to confirm (e.g. "Look good?") before proceeding to Phase 3. This prevents silent tool omissions. If the user corrects the list, update your selection accordingly.

Phase 3: Prompt Assembly

1. Generate a slug from the topic (lowercase, hyphens, max 40 chars)

- "review the auth flow" → auth-flow-review - "is this migration safe" → migration-safety-review

1. Create the output directory via Bash. The directory name MUST always be prefixed with a second-precision UNIX timestamp so runs are lexically sortable and never collide: ./agents/counselors/TIMESTAMP-[slug] For example: ./agents/counselors/1770676882-auth-flow-review Mac tip: Generate with date +%s (seconds since epoch). Millisecond precision is NOT available via date on macOS without GNU coreutils — use date +%s for portable second-precision timestamps.
2. Write the prompt file using the Write tool to ./agents/counselors/TIMESTAMP-[slug]/prompt.md:

# Review Request

## Question
[User's original prompt/question from $ARGUMENTS]

## Context

### Files Referenced
[Contents of the most relevant files found in Phase 1]

### Recent Changes
[git diff output, if any]

### Related Code
[Related files discovered via search]

## Reference Documentation
[Phase 1b content — current library docs from Context7. If Phase 1b was skipped, omit this section.]

## Instructions
You are providing an independent review. Be critical and thorough.
- Analyze the question in the context provided
- Identify risks, tradeoffs, and blind spots
- Suggest alternatives if you see better approaches
- Be direct and opinionated — don't hedge
- Structure your response with clear headings
- Flag any code patterns that appear outdated vs. the Reference Documentation above

Phase 4: Dispatch

Tell the user before dispatching:

"Dispatching to [N] agents: [list]. This typically takes 2-5 minutes..."

Note the prompt directory path you created (e.g. ./agents/counselors/1772865337-auth-flow-review/). The counselors CLI creates a sibling output directory with a second timestamp suffix.

Run counselors via Bash with the prompt file, passing the user's selected agents:

set -a; for f in ~/.env .env ~/.vibe-tools/.env; do [ -f "$f" ] && source "$f"; done; set +a; counselors run -f ./agents/counselors/[slug]/prompt.md --tools [comma-separated-selections] --json

Why the env sourcing? Claude Code's Bash tool may not inherit API keys (e.g. OPENAI_API_KEY) from the user's interactive shell. The set -a + source pattern loads keys from standard dotenv files portably (works in bash, zsh, sh). Files that don't exist are silently skipped.

Example: --tools claude,codex,gemini

Use Bash timeout: 480000 (8 minutes). Tools run in parallel (not sequentially). Per-tool timeouts in the counselors config control how long each individual tool gets.

Important: Use -f (file mode) so the prompt is sent as-is without wrapping. Use --json to get structured output for parsing.

Phase 5: Read Results (filesystem-based — does NOT depend on stdout)

IMPORTANT: Do NOT rely solely on JSON stdout. The CLI only writes run.json and prints JSON after ALL tools finish. If any tool hangs or the process is killed, stdout will be empty. Always fall back to scanning the filesystem.

Step 1: Find the output directory.

ls -dt ./agents/counselors/[slug]-*/ 2>/dev/null | head -1

If no directory found, the CLI failed before dispatching. Tell the user and suggest counselors doctor. Stop.

Step 2: Check for run.json (happy path). If run.json exists, parse it:

• status: "success" with wordCount > 0 — genuine success
• status: "timeout" — tool hit its timeout
• status: "error" — tool crashed
• status: "success" with wordCount: 0 — silent failure (read .stderr)

Step 3: If NO run.json, scan for individual files. For each expected tool, check if {tool-id}.md exists and has size > 0. Check {tool-id}.stderr for error details.

Step 4: Report to user.

• All tools produced output: Proceed to Phase 6.
• Some tools produced output: Tell the user which failed and why, then ask: "Continue with [N] of [M] responses, or retry?"
• Zero tools produced output: Report errors. Suggest counselors doctor. Stop.

Phase 6: Synthesize and Present

Combine all agent responses into a synthesis:

## Counselors Review

**Agents consulted:** [list of agents that responded]

**Consensus:** [What most agents agree on — key takeaways]

**Disagreements:** [Where they differ, and reasoning behind each position]

**Key Risks:** [Risks or concerns flagged by any agent]

**Blind Spots:** [Things none of the agents addressed that seem important]

**Recommendation:** [Your synthesized recommendation based on all inputs]

---
Reports saved to: [output directory from manifest]

Present this synthesis to the user. Be concise — the individual reports are saved for deep reading.

Phase 7: Action (Optional)

After presenting the synthesis, ask the user what they'd like to address. Offer the top 2-3 actionable items from the synthesis as options. If the user wants to act on findings, plan the implementation before making changes.

Error Handling

• counselors not installed: Tell the user to install it (npm install -g counselors)
• No tools configured: Tell the user to run counselors init or counselors add
• No output directory created: CLI failed before dispatching (bad config, missing binary). Check stderr from the Bash call.
• Output directory exists but no run.json: CLI was killed before all tools finished. Scan for individual .md files — completed tools will have written their output. This is the most common partial-failure mode.
• Silent failure (status: "success" but wordCount: 0, or .md file is 0 bytes): Read the .stderr file. Common causes: expired API key, 402 payment required, rate limit.
• Single agent fails: Note it, ask user whether to continue with remaining responses or retry.
• All agents fail: Report each error from .stderr files. Suggest counselors doctor. Do NOT proceed to synthesis.
• Never wait indefinitely: The 8-minute Bash timeout is the hard ceiling. Do not add sleep/retry loops.