meeting-recorder-assistant

Intelligent meeting recording and transcription assistant with automated minutes generation, action item extraction, and sentiment analysis. Supports audio transcription, speaker diarization, meeting summarization, and task extraction. Use when users need to record meetings, transcribe audio, generate meeting minutes, extract action items, or analyze meeting content.

Quality

93%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

87%

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

The body is concise and highly actionable with verified executable code and clean progressive disclosure to real bundle scripts. Its main weakness is workflow clarity: the record → transcribe → minutes → actions pipeline is presented as disconnected snippets with inconsistent transcript paths and no validation checkpoints tying the steps together.

Suggestions

Sequence the three examples into one end-to-end workflow showing each step's output feeding the next — e.g., explicitly save the recorder's transcript dict to /tmp/transcript.json before calling generate_minutes, since the current examples reference inconsistent transcript sources.

Add validation checkpoints between steps — e.g., check transcript['status'] == 'success' after recording and confirm the minutes dict contains expected keys before extracting action items — so failures are caught early.

Add brief error-recovery guidance for common failures (microphone unavailable, transcription API errors) so Claude knows how to handle the speech_recognition exception paths the scripts already return.

Dimension	Reasoning	Score
Conciseness	The body is lean and dominated by executable code; it does not explain concepts Claude already knows (e.g., what WAV is or how transcription works), matching the level-3 'lean and efficient; every token earns its place' anchor. The brief Features list mildly restates the description but stays tight rather than padded, so it does not drop to level 2.	3 / 3
Actionability	All three code examples use real imports and signatures that match the bundle scripts (verified: MeetingRecorder.start_recording/stop_and_transcribe, generate_minutes, extract_actions all exist with the shown keys), making them copy-paste ready — matching the level-3 'fully executable code; copy-paste ready' anchor.	3 / 3
Workflow Clarity	Three labeled subsections are present but they are disconnected as a workflow: stop_and_transcribe() returns a dict yet the minutes example reads from /tmp/transcript.json, with no step bridging them, and there are no validation/checkpoint steps — matching the level-2 'steps listed but validation gaps; checkpoints implicit' anchor. Not level 3 because no validation feedback loop exists; not level 1 because individual steps are clear and concrete.	2 / 3
Progressive Disclosure	For a single-purpose skill under ~65 lines, the body is well-organized into clear sections and references real one-level-deep bundle scripts (scripts/meeting_recorder.py, meeting_minutes.py, action_extractor.py) via imports, with no nested or broken references — matching the level-3 anchor and the simple-skill scoring note.	3 / 3
	Total	11 / 12 Passed

Description

100%

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

A strong, well-formed description: it states concrete capabilities in third person and provides an explicit 'Use when...' trigger clause with natural user-facing terms and a distinct meeting-focused niche. No significant weaknesses; it would surface correctly when needed and avoid conflicting with unrelated skills.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'automated minutes generation, action item extraction, and sentiment analysis' and 'audio transcription, speaker diarization, meeting summarization, and task extraction' — matching the level-3 anchor that lists several specific actions rather than a vague domain.	3 / 3
Completeness	Explicitly answers both what (recording, transcription, minutes, diarization, sentiment) and when ('Use when users need to record meetings, transcribe audio, generate meeting minutes, extract action items, or analyze meeting content'), matching the level-3 anchor with an explicit trigger clause; not level 2 because the 'when' is stated explicitly, not merely implied.	3 / 3
Trigger Term Quality	The trigger clause uses natural phrasing a user would actually say — 'record meetings, transcribe audio, generate meeting minutes, extract action items, or analyze meeting content' — giving good coverage rather than jargon-only terms.	3 / 3
Distinctiveness Conflict Risk	The meeting-recording/transcription niche is clearly scoped and all triggers are tied to meeting-specific scenarios, making it unlikely to fire for unrelated skills; matches the level-3 'clear niche with distinct triggers' anchor.	3 / 3
	Total	12 / 12 Passed

Validation

100%

Warnings & errors only

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

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: OpenBMB/PilotDeck
Path: skills/meeting-recorder-assistant/SKILL.md
Commit: c88d7d1

Reviewed: about 13 hours ago

Table of Contents

Discovery Implementation Validation

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.