one-on-one-prep — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited one-on-one-prep (Agent Skill) and scored it 96/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 1 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 1 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.
Great 1:1s live at the intersection of performance and care. Development keeps growth alive. Wellbeing makes sure the person behind the work is seen. Both matter every time.
Deep-dive preparation for 1:1 meetings with a specific direct report, anchored in the org's performance framework (from manager-context/performance-framework.md) and organizational values.
If any MCP connector is unavailable, follow the connector unavailability protocol in references/operating-principles.md.
If a name is provided, match it against manager-context/team/ profiles.
If no name is specified, check the calendar for the next upcoming 1:1 meeting and identify the attendee.
If ambiguous:
Which team member? Your upcoming 1:1s are:
- [Name], [day] at [time]
- [Name], [day] at [time]Load the team member's profile from manager-context/team/[name].md for:
If no profile exists:
⚠️ No profile found for [name]. Run /setup to build team context, or I'll work with what I can find from sources directly.Slack (last 7-14 days):
Google Drive:
Notion:
Compile into a "recent activity" summary.
Look for signals (NOT diagnoses, signals for the manager to explore):
Slack:
Calendar:
Important: Frame these as conversation starters, not assessments:
💡 Potential topics to explore (signals, not conclusions):
- [Name] asked about [topic] in #channel 3 times this week without a clear resolution, might be a blocker worth discussing
- [Name]'s activity in #project-channel has been lower than usual, could be fine, but worth checking inScan for signals related to the organization's values (from manager-context/values.md). These give the manager conversation threads beyond just goals and deliverables.
If no values are configured, focus on these universal management dimensions:
Wellbeing & Energy:
Team Connection & Collaboration:
Communication & Transparency:
Resourcefulness & Problem-Solving:
Ownership & Initiative:
If organizational values are configured, map these dimensions to the specific values and use value names in the output.
Compile into a brief values snapshot (not a scorecard, conversation prompts):
Values Snapshot:
- [Value/Dimension 1]: [observation or "no signal"]
- [Value/Dimension 2]: [observation or "no signal"]
..._Only include values/dimensions where there's a meaningful signal. Don't force all of them every time._
Using the goals location from the team member's profile:
Notion:
Google Drive:
Reference the org's performance framework dimensions (from manager-context/performance-framework.md). Use the org's dimension names and sub-dimensions when checking goal progress. If this file doesn't exist, ask the manager to run /setup first.
See the "Evidence Gathering Guidelines" section in references/performance-framework.md for what to look for per dimension type:
📋 Development Goal Status:
- Goal 1: "[goal text]", [status: on track / needs attention / no update since [date]]
- Goal 2: "[goal text]", [status]
- Development area: "[area]", [any evidence of progress or attention]
⏰ Last goal update: [date], [flag if >6 weeks old]Search for the last 1:1 document/notes:
Google Drive: Search "[name] 1:1", "one on one [name]"
Extract:
📝 From last 1:1 ([date]):
Open items:
- [ ] [Manager] to [action], [status: done/pending/unknown]
- [ ] [Team member] to [action], [status: done/pending/unknown]
Parked topics: [if any]Read references/output-template.md for the full output template structure.
Spawn a sub-agent to review the prep sheet with fresh eyes. The reviewer should:
Incorporate the reviewer's feedback before presenting the final prep sheet.
Here's your prep for your 1:1 with [name]. Anything you'd like me to dig deeper on?Read references/operating-principles.md for shared operating principles (data scope, DM flagging, signals vs diagnoses, connector unavailability).
Additional notes specific to this skill:
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.