Skillv3.4.2

ClawScan security

Portfolio · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

BenignMar 18, 2026, 10:49 AM

Verdict: benign
Confidence: high
Model: gpt-5-mini
Summary: The skill's code and instructions are consistent with a local portfolio manager that stores data under ~/.portfolio and does not request external credentials or network access.
Guidance: This skill appears to do exactly what it says: a local portfolio manager that writes/reads ~/.portfolio/holdings.json and transactions.json and uses python3 (standard library). Before installing or running: (1) inspect the script (already included) and confirm you are comfortable with data being stored unencrypted in ~/.portfolio; (2) back up or move any existing ~/.portfolio files to avoid accidental overwrite; (3) note the small feature mismatch (analyze --by sector is documented but not implemented); and (4) if you plan to store sensitive financial records, consider filesystem permissions or encryption for the data directory. If you want absolute isolation, run the script in a container or limited user account.

Purpose & Capability: noteThe name/description (portfolio manager) aligns with the included script and SKILL.md: commands for add/remove/list/analyze/rebalance/performance are implemented. Minor mismatch: SKILL.md documents an analyze --by ticker|sector flag, but the script's analyze implementation always aggregates by ticker (sector grouping is not implemented). Otherwise the capabilities requested are proportional to the stated purpose.
Instruction Scope: okRuntime instructions and the script operate only on local files in ~/.portfolio (holdings.json and transactions.json) and print to stdout. The SKILL.md claims no external API calls and the script does not perform network I/O or attempt to read unrelated system files or credentials.
Install Mechanism: okNo install spec is provided (instruction-only packaging) and the shipped script is executed locally. There are no downloads or third-party package installs; risk from install mechanism is low.
Credentials: okThe skill declares no required environment variables or credentials. At runtime it uses HOME for the default data directory and passes its own variables into embedded Python blocks — there are no requests for secrets or unrelated credentials.
Persistence & Privilege: okalways:false and model invocation is normal. The skill persists only its own data to ~/.portfolio and does not modify other skills, system configs, or request elevated privileges.