Graph Of Thoughts
PassAudited by ClawScan on May 10, 2026.
Overview
This is mostly a reasoning-template skill with no credentials or network access, but it includes optional persistent memory guidance and a wrapper that tries to load a missing local implementation file.
This skill appears safe for ordinary reasoning use. Before installing, be aware that its runnable wrapper references a missing local implementation file, and avoid enabling or following the memory-recording guidance for confidential problems unless you intentionally want those details saved.
Findings (2)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
The skill may fail to run as packaged, or could execute a local implementation file if one is later added in that directory.
The wrapper dynamically loads a local implementation file that is not included in the manifest, so any runnable implementation is not fully represented in the reviewed artifacts.
spec = importlib.util.spec_from_file_location("graph_of_thoughts", Path(__file__).parent / "graph_of_thoughts.py") ... spec.loader.exec_module(module)Include the referenced implementation file in the package or remove the wrapper; if dynamic loading is required, clearly document the file provenance and expected behavior.
If users apply this to sensitive work, problem details or decisions could remain available to future sessions.
The documentation encourages retaining problem statements, reasoning sessions, and derived patterns in persistent memory.
When you successfully use GoT on a new problem: 1. **Document the session** in `memory/got-sessions.md` ... 3. **Extract patterns** for future reference
Store session memory only with user consent, redact sensitive details, and define retention or deletion expectations.