meeting-transcript

Process meeting recordings and notes into structured decisions, action items, and team dynamics with intelligent noise filtering

Quality

47%

Does it follow best practices?

Impact

—

No eval scenarios have been run

Securityby

Advisory

Suggest reviewing before use

Optimize this skill with Tessl

npx tessl skill review --optimize ./.claude/skills/meeting-transcript/SKILL.md

Quality

Content

27%

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

This skill covers meeting transcript processing comprehensively but suffers from significant verbosity, explaining many concepts Claude already understands (what side conversations are, what action items look like). The parallel agent strategy is a useful structural contribution, but lacks concrete invocation syntax and validation checkpoints. The entire document would benefit from being split into a concise overview SKILL.md with supporting reference files for templates, filtering rules, and domain-specific processing.

Suggestions

Cut the skill by at least 50%: remove the INCLUDE/EXCLUDE examples (Claude knows what side conversations are), collapse the Content Filtering Protocol into the Phase 1 setup, and trim the Content Structure to just field names without descriptions of obvious fields.

Add concrete Task tool invocation syntax showing exactly how to spawn agents with run_in_background: true, rather than just describing the concept.

Add a validation checkpoint in Phase 3 before saving—e.g., verify action items have owners, decisions have rationale, and flag any agent that returned empty results.

Split into multiple files: move the metadata template, content structure template, and content filtering guidelines into separate referenced files to improve progressive disclosure.

Dimension	Reasoning	Score
Conciseness	The skill is extremely verbose at ~200+ lines with significant redundancy. The content filtering guidelines explain obvious concepts (what side conversations are, what technical difficulties are). The Content Filtering Protocol repeats filtering concepts already covered. The Learning Integration section is vague filler. Much of the Content Structure section describes obvious fields Claude would know to include.	1 / 3
Actionability	The parallel agent strategy provides concrete agent prompts and the metadata template is copy-paste ready YAML. However, there's no executable code, the agent spawning instructions reference a 'Task tool' without showing exact invocation syntax, file paths are all placeholder '[CUSTOMIZE: path]', and much of the guidance is descriptive rather than executable.	2 / 3
Workflow Clarity	The three-phase workflow (Setup → Parallel Processing → Assembly) is clearly sequenced and the solo vs team mode branching is a nice touch. However, there are no validation checkpoints—no step to verify agent results before assembly, no error handling if an agent fails, and no feedback loop for when transcript quality is poor or results are contradictory.	2 / 3
Progressive Disclosure	This is a monolithic wall of text with no references to external files despite being long enough to warrant splitting. The content filtering guidelines, metadata template, content structure template, and domain-specific processing rules could all be separate referenced files. No bundle files exist to offload this content to.	1 / 3
	Total	6 / 12 Passed

Description

67%

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 does a good job specifying concrete capabilities and carving out a distinct niche around meeting processing. However, it lacks an explicit 'Use when...' clause with trigger terms, which weakens its completeness and discoverability. Adding natural trigger keywords and explicit usage guidance would significantly improve skill selection accuracy.

Suggestions

Add a 'Use when...' clause such as 'Use when the user asks to summarize meetings, extract action items, process meeting transcripts, or review meeting minutes.'

Include additional natural trigger terms users might say: 'meeting minutes', 'transcript', 'meeting summary', 'follow-ups', 'takeaways', 'standup notes'.

Dimension	Reasoning	Score
Specificity	Lists multiple specific concrete actions: 'structured decisions, action items, and team dynamics' with 'intelligent noise filtering'. These are concrete, identifiable outputs rather than vague abstractions.	3 / 3
Completeness	Clearly answers 'what does this do' (processes meeting recordings/notes into structured outputs), but lacks an explicit 'Use when...' clause or equivalent trigger guidance, which caps this at 2 per the rubric.	2 / 3
Trigger Term Quality	Includes natural terms like 'meeting recordings', 'notes', 'decisions', 'action items', but misses common variations users might say such as 'meeting minutes', 'meeting summary', 'transcript', 'takeaways', or 'follow-ups'.	2 / 3
Distinctiveness Conflict Risk	The combination of 'meeting recordings and notes' with specific outputs like 'decisions, action items, and team dynamics' creates a clear niche that is unlikely to conflict with other skills like general note-taking or document processing.	3 / 3
	Total	10 / 12 Passed

Validation

90%

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 — 10 / 11 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	10 / 11 Passed

Repository: huytieu/COG-second-brain
Commit: 034af4c

Reviewed: about 1 month 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.