Decision Topology
v1.0.6Records the structure of conversations where ideas evolve, branch, get rejected, pivot, or combine. Saves each structural shift as a node in a local JSON tre...
⭐ 0· 371·0 current·0 all-time
byJounes De Schrijver@joncik91
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description (record conversation structure) matches the actual footprint: a Node.js script that reads stdin, updates local JSON trees, writes companions, and maintains a local concept index. Required binary 'node' is appropriate and no unrelated credentials or services are requested.
Instruction Scope
SKILL.md explicitly restricts operations to stdin input, file writes inside the trees directory, and no network calls; the included script enforces truncation, path containment, and uses only fs/path. The one point to watch: the skill generates companion .md files and a global concepts.json index (for semantic/search use) — these files contain truncated summaries/reasoning and could be visible to other local services if you point the trees dir at a shared/indexed location.
Install Mechanism
No install spec; this is an instruction-and-code skill bundled with its script files. There are no remote downloads, package installs, or unusual installers — only a Node.js script included in the skill bundle.
Credentials
No secrets or extraneous environment variables are required. Optional override TOPOLOGY_TREES_DIR is reasonable for storage configuration; the metadata requires only the 'node' binary. The suggested use of a memory directory for indexing is user-controlled and not mandatory.
Persistence & Privilege
The skill sets always: true in its metadata (active by default). That is consistent with the described 'low-noise, always-on' behavior, but it's a meaningful privilege — the skill will autonomously create and update local files during conversations unless you change metadata to always: false or disable it. The script does not modify other skills or global agent configs.
Assessment
This skill appears to do what it says: it logs conversation structure to local JSON files and companion .md files and enforces truncation and path containment. Before installing: (1) Be aware it is active by default (always: true) — set always: false in SKILL.md if you prefer explicit invocation. (2) Inspect or sandbox the storage location: the default {baseDir}/trees/ will contain JSON and .md files plus concepts.json; do not point TOPOLOGY_TREES_DIR at a directory scanned or uploaded by other services unless you want that data indexed. (3) If you handle sensitive conversations, test the skill with a temporary directory (TOPOLOGY_TREES_DIR=/tmp/test-trees) to confirm behavior and review the files it creates. (4) You can review scripts/topology.js yourself — it uses only fs/path, truncates fields, enforces safe paths, and has no network/child_process calls. (5) If you need stricter guarantees, change always to false and enable the skill only when you want structured logging.Like a lobster shell, security has layers — review code before you run it.
cognitivevk972pjctsdbp1w50gk1yz5tf5n825gvfconversationvk972pjctsdbp1w50gk1yz5tf5n825gvflatestvk97dtswff4t0gp0fdy2j6341ah824gbwmemoryvk972pjctsdbp1w50gk1yz5tf5n825gvftrackingvk972pjctsdbp1w50gk1yz5tf5n825gvf
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
Runtime requirements
🌳 Clawdis
Binsnode