keep-learning

v0.0.2

Learn and memorize knowledge from local directories. Supports Markdown and code files. Extracts key insights, builds knowledge index, and stores in agent mem...

1· 293·2 current·2 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
Purpose & Capability
Name/description claim: learn from local directories and store memories. Snack: SKILL.md only asks for filesystem access, optional git, and agent memory operations — all coherent with that goal. No unrelated env vars, binaries, or remote-only credentials are requested.
Instruction Scope
Instructions explicitly require reading files, invoking git (git pull/diff/status) when applicable, using read_file/update_memory/search_memory, and writing a small runtime directory (~/.keep-learning). This is consistent with purpose, but the skill will read arbitrary user files under the provided path and may perform network activity via git pull — a privacy/telemetry consideration the user should be aware of.
Install Mechanism
Instruction-only skill with no install spec and no code files. Lowest installation risk; nothing is downloaded or written besides its own runtime state (~/.keep-learning) as documented.
Credentials
No environment variables, credentials, or unrelated config paths are requested. The skill's needs (filesystem access, optional git, agent memory APIs) are proportional to its functionality.
Persistence & Privilege
always:false (normal). The skill writes its own runtime files to ~/.keep-learning and creates/updates entries in the agent memory system — expected for a learning/memory skill. Users should note it will persist learned summaries and last-commit state locally.
Assessment
This skill is coherent for indexing and memorizing local knowledge, but it necessarily reads all files under the user-provided path and may run git pull (network access) if the folder is a repo. Before installing or enabling: 1) Only point the skill at directories you are comfortable having the agent read and summarize. 2) Be aware the skill will store summaries and a last-commit file in ~/.keep-learning. 3) If you do not want network accesses, avoid using it on git repositories or disable auto-pull in your workflow. 4) Confirm the agent has appropriate permissions and review memory entries periodically to remove any sensitive items the skill may have stored.

Like a lobster shell, security has layers — review code before you run it.

latestvk97eqcmq50493k8yzevecpw9z183fgqn

License

MIT-0
Free to use, modify, and redistribute. No attribution required.

Comments