Organize your agent's knowledge using PARA (Projects, Areas, Resources, Archive) — then make it ALL searchable. The symlink trick enables full semantic search across your entire knowledge base, not just MEMORY.md. Includes session transcript indexing and memory flush protocol. Your agent finally has a real second brain.
Security Analysis
high confidenceThe skill's files and runtime instructions are consistent with a file-based PARA second-brain: it only scaffolds directories, writes templates, and recommends a symlink and config change to expand search scope — nothing requested or installed is disproportionate to that purpose.
The name/description match the actions in the repo: SKILL.md explains making notes searchable, templates provide PARA content, and scripts/setup.sh scaffolds the workspace and creates MEMORY.md and today's log. No extraneous credentials, downloads, or unrelated binaries are requested.
Instructions include creating a symlink from your notes/ into memory/ (so memory_search indexes all notes) and changing the Clawdbot memorySearch config to add sessions. These steps are coherent for the stated goal but intentionally broaden what the agent will index (including session transcripts). That expansion can expose sensitive contents if you symlink arbitrary system paths or have secrets in notes.
There is no install spec (instruction-only). The included scripts/setup.sh is a small, local scaffolding script that only creates directories and writes template files in the provided workspace path; it performs no network downloads or extraction.
The skill requests no environment variables or credentials. However, because it expands indexing to entire notes/ and session transcripts, any secrets already present in files (API keys, tokens, config snippets) could become searchable and effectively exposed to the agent's memory_search results.
always is false and the skill does not attempt to modify other skills or global agent settings beyond recommending an edit to the agent's memorySearch config and AGENTS.md. The default ability for the model to invoke the skill autonomously is unchanged (platform default) and not combined with other red flags here.
Guidance
This appears to do what it says, but review before enabling: 1) Inspect scripts/setup.sh (it only creates folders and templates). 2) Do NOT ln -s arbitrary system directories into memory/ — only point the symlink at the intended workspace/notes folder; symlinking sensitive paths could expose system files to indexing. 3) Before enabling session transcript indexing or symlinking, search your notes for any secrets (API keys, tokens, passwords) and remove or exclude them. 4) Back up current MEMORY.md and AGENTS.md, then test in an isolated workspace. 5) If you later want to limit what gets indexed, remove the symlink and revert the memorySearch config change. If you want more assurance, ask the skill author for a canonical source/homepage or a signed release to verify provenance.
Latest Release
v2.0.1
Added PARA explanation to description, credited Tiago Forte as creator of PARA methodology
More by @halthelobster
Published by @halthelobster on ClawHub
