CtrlK
BlogDocsLog inGet started
Tessl Logo

minutes-live-sidekick

Act as the user's live meeting sidekick inside the current terminal agent session. Use when the user explicitly asks you, the terminal agent, to watch a meeting, follow the live transcript, answer during the call, offer strategist thoughts, silently watch for risks, or track decisions. Do not use this skill to start or control the separate Minutes Coach HUD; explicit Coach or HUD lifecycle requests belong to minutes-copilot, and an ambiguous request such as "coach me live" requires one short surface clarification.

72

Quality

89%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Quality

Content

78%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is an instruction-rich, well-structured skill with concrete commands and a validated closing workflow. It is mostly lean and fully navigable, with only minor conciseness and mid-flow checkpoint gaps.

Suggestions

Tighten the 'Respect foreground priority' and 'Treat meeting context as evidence' sections by merging the bulleted 'Never' list with the restated proactive-monitoring paragraph to remove duplicated prohibitions.

Add explicit validation checkpoints (e.g., 'confirm capture state before attaching' / 're-confirm posture after a role change') to the mid-flow sections so the whole sequence matches the checklisted rigor of the End-and-hand-off section.

Where the adapter is left open ('Prefer a documented exact-session event or wait adapter when the host exposes one'), either name the supported adapter or state plainly that none is currently available, so the guidance is copy-paste ready instead of conditional.

DimensionReasoningScore

Conciseness

Dense, rule-oriented prose that assumes Claude's competence without explaining basic concepts, but the foreground-priority and treat-as-evidence sections restate the same prohibitions, leaving minor trim opportunities. Fits anchor-4 (efficient with minor over-explanation) better than the lean anchor-5.

4 / 5

Actionability

Gives concrete commands (minutes transcript --status, --since 2m, --since <cursor>, minutes stop, minutes status) and a verbatim disambiguation question, but defers the event/wait adapter to 'a documented ... adapter when the host exposes one', a minor gap keeping it below anchor-5.

4 / 5

Workflow Clarity

A clear surface-to-handoff sequence exists and the End-and-hand-off section is an explicit, validated checklist with state-reporting and finalization gates, but only that closing section is fully checklisted; mid-flow transitions lack explicit checkpoints, so it sits at anchor-4 rather than 5.

4 / 5

Progressive Disclosure

Well-organized with clear section headers, self-contained single-file structure appropriate for a guidance skill, and one-level-deep, clearly signaled references to sibling skills (/minutes-copilot, /minutes-debrief) with no nested file indirection — matching the anchor-5 example.

5 / 5

Total

17

/

20

Passed

Description

100%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is specific, trigger-rich, complete, and strongly distinct from the sibling minutes-copilot skill. It uses imperative third-person voice with no first/second-person drift and includes an explicit negative boundary plus an ambiguity-resolution trigger.

DimensionReasoningScore

Specificity

Lists multiple concrete actions — 'watch a meeting, follow the live transcript, answer during the call, offer strategist thoughts, silently watch for risks, or track decisions' — giving comprehensive coverage comparable to the anchor-5 example.

5 / 5

Completeness

Explicitly answers both what (act as live sidekick with enumerated postures) and when ('Use when the user explicitly asks you, the terminal agent, to watch a meeting...'), with concrete trigger phrases per the anchor-5 example.

5 / 5

Trigger Term Quality

Captures natural user phrasings — 'watch a meeting', 'follow the live transcript', 'answer during the call', 'coach me live' — including the ambiguous disambiguation case, matching comprehensive anchor-5 coverage.

5 / 5

Distinctiveness Conflict Risk

Carves a clear niche distinct from minutes-copilot via the terminal-agent-vs-Coach-HUD boundary and the 'coach me live' clarification, yielding minimal conflict risk as in the anchor-5 example.

5 / 5

Total

20

/

20

Passed

Validation

100%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
silverstein/minutes
Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.