Weclone Twin Reply
PassAudited by ClawScan on May 1, 2026.
Overview
The skill coherently drafts review-gated persona-style replies, but users should understand it reads and packages sensitive persona and conversation files for a model prompt.
This skill appears safe for its stated purpose if you are comfortable sharing the selected persona files and conversation window with the model used for drafting. Keep the review gate, do not auto-send drafts, and make sure the `ai_twin/` directory contains only information you intend to use for this persona.
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.
Personal persona details and recent conversation context may be included in the rendered prompt used to generate a reply.
The skill intentionally relies on persistent persona and state files that may contain private identity, values, and behavioral information.
Default to `ai_twin/` at the repo root, containing `profile.md`, `persona_examples.md`, `guardrails.md`, and optional `state.md`.
Keep the persona directory curated, avoid storing secrets or unrelated private notes there, and review the rendered prompt before using it with a model.
Unintended or outdated Markdown files in the persona folder could affect the draft or expose extra context in the prompt.
The renderer automatically includes additional Markdown files in the persona directory, so extra files placed there can influence the generated prompt.
extras = sorted(
path
for path in persona_dir.glob("*.md")
if path.name not in known_names and path.name not in excluded_names
)Store only intended persona files in the persona directory, and remove or relocate unrelated Markdown files before rendering a prompt.