CtrlK
BlogDocsLog inGet started
Tessl Logo

jbaruch/nanoclaw-core

Core behavioral rules and skills for NanoClaw personal assistant agents. Always-on rules for communication, verification, memory, and formatting.

97

Quality

97%

Does it follow best practices?

Impact

Pending

No eval scenarios have been run

SecuritybySnyk

Advisory

Suggest reviewing before use

Overview
Quality
Evals
Security
Files

progress-updates.mdrules/

alwaysApply:
Yes

Progress Updates for Long Tasks

When launching a background agent for a task that may take more than 30 seconds, include this instruction in the agent prompt:

If your work takes more than 30 seconds, send a brief progress update via mcp__nanoclaw__send_message before delivering the final result. Keep it short and contextual — what you're doing right now, not a status report.

Examples of good progress updates

  • "Ищу в почте за последние 3 дня..."
  • "Нашёл 4 письма, анализирую..."
  • "Checking 3 GitHub repos..."
  • "Found the issue, writing fix..."

Examples of bad progress updates

  • "Still working on your request." (generic, no context)
  • "Processing step 2 of 5..." (robotic)
  • "I'm currently in the process of..." (LLM-speak)

Rules

  • The initial emoji reaction remains the first acknowledgement — progress updates come AFTER, only if work is lengthy.
  • One update is usually enough. Two max for very long tasks. Don't spam.
  • Match the language of the original message.
  • If the task finishes quickly (<30s), no update needed — just deliver the result.
  • Progress updates use send_message with the same reply_to as the final result.

rules

context-recovery.md

core-behavior.md

default-silence.md

ground-truth.md

language-matching.md

post-compaction-trust.md

progress-updates.md

query-size-limits.md

read-full-content.md

telegram-protocol.md

temporal-awareness.md

tone-matching.md

README.md

tile.json