using-exploration-cycle — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited using-exploration-cycle (Agent Skill) and scored it 87/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 3 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 3 flagged
The text {match} tells the agent to skip the normal "ask the user first" gate. Used adversarially it removes the human-in-the-loop check before destructive or sensitive actions, turning a normally-gated agent into a fire-and-forget executor.
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
<SUBAGENT-STOP> If you were dispatched as a subagent (via dispatch.py) to execute a specific task, skip this skill. </SUBAGENT-STOP>
<EXTREMELY-IMPORTANT> You are operating with the exploration-cycle-plugin active.
Before responding to ANY user message (including answering general questions or scoping requests), you MUST verify the state of the active exploration session. If a session is in progress, you DO NOT have a choice: you MUST route execution through the active phase of the exploration-workflow. </EXTREMELY-IMPORTANT>
Before generating any response:
exploration/exploration-dashboard.md exists.**Status:** field.Complete, the prior session has ended. You may answer normally or offer to start a new exploration.In Progress or TBD, identify the **Current Phase:**.exploration-workflow skill. Never directly invoke child skills unless authorized by the orchestrator. Do NOT answer the user's question directly in freeform prose.exploration-workflow skill to initiate Phase 0 intake.The SQLite state database and programmatic phase artifacts are the absolute state authority. The markdown dashboard is a read-only projection of the database. Do not rely on conversational chat history to assume a phase is complete or that a gate has been passed.
When executing the workflow or waiting for sub-agent execution / task completions:
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.