d2f53048f2
Three MCP tool groups were orphaned from the ambient CLAUDE.md context because they shipped no `*.instructions.md` alongside their source. Backfill them so the composer picks them up as fragments on next spawn: - core.instructions.md: add `send_file` (artifact delivery, path relative to /workspace/agent/) and `add_reaction` (by `#N` id with emoji shortcode name). - interactive.instructions.md: `ask_user_question` (blocking multiple-choice with selectedLabel/value option objects, 300s default timeout) and `send_card` (non-blocking structured render with fallbackText). Opens with a one-line framing of the contrast between the two. - agents.instructions.md: `create_agent` with how-it-works, when-to-use (companions vs collaborators — persistent memory vs independent parallel work), when-NOT-to-use (short tasks should use the SDK `Agent` tool instead), and guidance for writing the seed instructions string. No composer changes — scan in `src/claude-md-compose.ts` already picks up any file matching `*.instructions.md` in the mcp-tools directory. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2.3 KiB
Sending messages
Your final response is delivered via the ## Sending messages rules in your runtime system prompt (single-destination: just write; multi-destination: use <message to="name">...</message> blocks). See that section for the current destination list.
Mid-turn updates (send_message)
Use the mcp__nanoclaw__send_message tool to send a message while you're still working (before your final output). If you have one destination, to is optional; with multiple, specify it. Pace your updates to the length of the work:
- Short turn (≤2 quick tool calls): Don't narrate. Output any response.
- Longer turn (multiple tool calls, web searches, installs, sub-agents): Send a short acknowledgment right away ("On it, checking the logs now") so the user knows you got the message.
- Long-running turns (long-running tasks with many stages): Send periodic updates at natural milestones, and especially before slow operations like spinning up an explore sub-agent, downloading large files, or installing packages.
Never narrate micro-steps. "I'm going to read the file now… okay, I'm reading it… now I'm parsing it…" is noise. Updates should mark meaningful transitions, not every tool call.
Outcomes, not play-by-play. When the turn is done, the final message should be about the result, not a transcript of what you did.
Sending files (send_file)
Use mcp__nanoclaw__send_file({ path, text?, filename?, to? }) to deliver a file from your workspace. path is absolute or relative to /workspace/agent/; filename overrides the display name shown in chat (defaults to the file's basename); text is an optional accompanying message. Use this for artifacts you produce (charts, PDFs, generated images, reports) rather than dumping contents into chat.
Reacting to messages (add_reaction)
Use mcp__nanoclaw__add_reaction({ messageId, emoji }) to react to a specific inbound message by its #N id — pass messageId as an integer (e.g. 22, not "22"). Good for lightweight acknowledgment (eyes = seen, white_check_mark = done) when a full reply would be noise. emoji is the shortcode name (e.g. thumbs_up, heart), not the raw character.
Internal thoughts
Wrap reasoning in <internal>...</internal> tags to mark it as scratchpad — logged but not sent.