consensus-persona-generator
PassAudited by VirusTotal on May 11, 2026.
Overview
Type: OpenClaw Skill Name: consensus-persona-generator Version: 1.1.14 The skill bundle is classified as benign. It adheres to its stated purpose of generating and persisting persona sets locally, without making any network calls (verified by `src/llm.mjs` using static data). Robust input validation is implemented in `src/validate.mjs`, mitigating common injection risks. File system operations are confined to reading a specified input file and writing artifacts to a local 'out' directory or a configured state path, with `README.md` explicitly warning against misconfiguring sensitive environment variables like `CONSENSUS_STATE_ROOT`. There is no evidence of data exfiltration, malicious execution, persistence mechanisms, or prompt injection attempts against the agent.
Findings (0)
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.
Running the skill can create or update local consensus board artifacts that later workflows may read.
The skill intentionally mutates local board state by writing a persona_set artifact. This is disclosed and purpose-aligned, but users should notice that invocation changes persistent consensus state.
const write = await writeArtifact(board_id, 'persona_set', payload, statePath);
Use a dedicated, non-privileged consensus state directory and review generated persona_set artifacts if they will affect important decisions.
Saved personas may shape downstream guard or consensus decisions until regenerated or removed.
The skill deliberately creates durable persona context and may reuse it across later consensus workflows. This is central to the purpose, but persistent context can become stale or influence future decisions.
persists a versioned `persona_set` artifact to board state ... reuses compatible persona sets when possible
Regenerate persona sets when the domain, risk profile, or governance assumptions change, and keep consensus state separate from unrelated private data.