memory-capture_skill

This skill captures memory with compact session or quick tips, routing input to structured exports or rapid notes.
  • TypeScript

1.3k

GitHub Stars

1

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill catlog22/claude-code-workflow --skill memory-capture

  • SKILL.md5.0 KB

Overview

This skill provides unified memory capture with intelligent routing between two modes: Compact for full-session preservation and Tips for quick note-taking. It detects user intent from keywords, flags, or input length and routes the request to the appropriate phase, then saves structured output to the shared memory store. The design prioritizes fidelity in Compact mode and speed in Tips mode.

How this skill works

The router parses input and applies auto-route rules (keywords, flags, length heuristics) to select either Compact or Tips. Compact produces a verbatim, structured export of the current session for recovery; Tips creates a short, tagged note with optional contextual metadata. Both modes persist their generated structured text via a core memory import API and return a confirmation ID or error guidance.

When to use it

  • You want a complete, recoverable snapshot of an active session (Compact).
  • You need a fast note, idea, or snippet saved with tags (Tips).
  • You have short text or flags like --tag or --context that indicate a quick memo.
  • You need to preserve session state for later recovery or handoff.
  • Input is ambiguous—use the interactive prompt to choose the correct mode.

Best practices

  • Use Compact only when you want the full session recorded verbatim for future restoration.
  • Prefer Tips for short notes under ~100 characters or when using --tag/--context flags.
  • Provide explicit tags and context with tips to improve searchability and reuse.
  • If routing is ambiguous, follow the skill's ask prompt rather than forcing a mode.
  • Rely on the returned confirmation ID for retrieval and cross-referencing.

Example use cases

  • Compact the current engineering discussion before ending a long design session for later recovery.
  • Quickly save a config suggestion: /memory-capture tip "Use Redis for rate limiting" --tag config.
  • Capture an important multi-step debugging plan verbatim so another agent can resume it.
  • Record a short insight or reminder while coding, with context pointing to the related feature.
  • When uncertain, invoke the skill without args to receive a mode-selection prompt.

FAQ

The skill retries once; if that fails it returns the generated structured text and an error message so you can save it manually.

How does the router decide between Compact and Tips?

It uses keyword matches, flags (--tag/--context), input length heuristics, and an interactive prompt for ambiguous cases.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational
memory-capture skill by catlog22/claude-code-workflow | VeilStrat