replay — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited replay (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
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.
This skill spawns a parallel "what-if" timeline rooted at an existing mem_snapshots row. It does not modify mem_nodes, mem_bt_ratings, or mem_judgements. The branch is recorded in mem_replay_branches and surfaced in the cockpit's Replay tab.
/replay.branch_paused event the user disagrees with — replay lets them simulate the alternative path before deciding whether to call mcp__memory__resume_branch.mcp__memory__snapshot outputs in earlier turns)."Approve hyp_07 instead of pruning it""Tighten preregistration alpha to 0.01 instead of 0.05""Use seed budget 10 instead of 3 for hyp_03"mcp__memory__replay_counterfactual(snapshot_id, counterfactual). Capture the returned branch_id.branch_id so the cockpit can group them.cockpit_interventions row (kind=note) so the user can promote it later via mcp__memory__resume_branch. Do not flip statuses yourself.mem_nodes from this path; you should refuse too.branch_ids.seed_perturb, query_heldout, or any model script. It is purely an inspection over already-recorded state.mcp__memory__list_replay_branches(limit=N).~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.