conversation-distill
v1.0.0At the natural end of a conversation, proactively suggest a structured wrap-up (distill): scan the full session, classify outputs into 6 categories (insights...
⭐ 0· 58·0 current·0 all-time
byYING99@yiing99
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
The name/description (session distillation into categories + write-to-notes) matches the SKILL.md and README. The skill does not declare or require unrelated credentials, binaries, or access. References to KnowMine or other MCPs are reasonable for a notes-writing feature and are presented as optional integrations.
Instruction Scope
Runtime instructions stay within scope: scan the conversation (which the agent already has access to), classify into the six categories, require explicit user confirmation before any write, and then write to the user's chosen notes tool. The instructions do not ask the agent to read system files, undisclosed env vars, or to transmit data to unexpected endpoints without user confirmation. The only external interactions are with user-configured notes backends (KnowMine, Notion, Obsidian, etc.), which is appropriate for the stated purpose.
Install Mechanism
This is an instruction-only skill (no install spec, no code files). The README mentions optional install commands (npx clawhub, GitHub plugin) for convenience, but the skill itself doesn't require installing arbitrary binaries or fetching/executing remote archives during runtime.
Credentials
The skill declares no required environment variables, primary credential, or config paths. It does reference that it will use whatever notes MCP the user has configured — any credentials needed to authorize those tools would come from the user's existing MCP setup, not from the skill asking for unrelated secrets.
Persistence & Privilege
Flags are default (always:false, user-invocable:true). The skill does not request persistent elevated presence or modify other skills' configurations. It explicitly enforces a mandatory confirmation step before writing, which limits autonomous writes.
Assessment
This skill appears internally consistent and behaves as described, but before installing: (1) verify which notes backend (KnowMine, Notion, Obsidian, etc.) you plan to use and confirm its permissions — granting write access to a third-party notes service is normal but should be given only to trusted tools; (2) confirm that your MCP integration requires explicit user consent for writes (the skill enforces a "write" confirmation, but double-check your backend's auth/consent flow); (3) be cautious about installing optional add-ons referenced in the README (npx/clawhub or GitHub plugin) and only install from sources you trust; and (4) if you handle sensitive data in conversations, remember that writing to external systems creates a new storage copy — review retention and access controls for the destination.Like a lobster shell, security has layers — review code before you run it.
latestvk973qa4a60a9yzdkh8qzav18ax84q73v
License
MIT-0
Free to use, modify, and redistribute. No attribution required.