advisor-update-writer — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited advisor-update-writer (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.
Write concise research updates that help an advisor, mentor, collaborator, or lab audience make decisions.
Use this skill when:
Do not use this skill for a full experiment report. Use experiment-report-writer when the main artifact is a detailed technical report. Use this skill when the main artifact is a decision-oriented communication.
Pair this skill with:
research-project-memory to recover current status, decisions, claims, evidence, risks, and actionsexperiment-report-writer when detailed experiment evidence needs a source reportfigure-results-review when plots or tables will be shown to an advisorresult-diagnosis when negative or ambiguous results need a decisionpaper-positioning-planner when the update asks for paper-story or target-venue feedbackpaper-evidence-board when writing and experiment gaps need a synchronized action listwork-timeline-planner when the update needs retrospective timeline evidence<installed-skill-dir>/
├── SKILL.md
└── references/
├── decision-framing.md
├── memory-writeback.md
└── update-template.mdreferences/decision-framing.md and references/update-template.md before drafting a saved or high-stakes update.references/memory-writeback.md when the update creates decisions, actions, or durable feedback that should persist.Choose the mode:
weekly: progress since last update, current blockers, next week plandecision: options, evidence, recommendation, explicit askmeeting: agenda, discussion points, decisions needed, notes after meetingemail: concise message with context, status, ask, and attachmentslab: broader context, key result, what the audience should rememberIdentify:
If saving and no path is given, use:
docs/updates/advisor_update_YYYY-MM-DD.mdPrefer project memory when available, then primary files.
Look for:
memory/current-status.mdmemory/decision-log.mdmemory/claim-board.mdmemory/evidence-board.mdmemory/risk-board.mdmemory/action-board.mdExtract:
Read references/decision-framing.md.
For each decision, produce:
If there is no decision needed, frame the update around progress, risks, and next actions.
Read references/update-template.md.
Default structure:
For email, make the first paragraph enough to answer "what do you need from me?"
For meeting notes, include both agenda and post-meeting decisions if the meeting has already happened.
Before finalizing:
Read references/memory-writeback.md when memory exists.
Update decisions, actions, risks, and status after the advisor responds or after meeting notes are finalized.
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.