CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

meeting-ingestion

Ingest meeting transcripts from ANY meeting recorder into brain pages with attendee enrichment, entity propagation, and timeline merge. One unified pipeline: normalize the source into a standard transcript record, split multi-meeting recordings, resolve speakers by evidence, create the page, pass every surprising claim through the consistency check (transcript + brain + plausibility), enrich every entity, then run the verification checklist — substance AND sequence. A meeting is NOT fully ingested until the enrich skill has processed every entity AND the verification checklist passes, including the sequence verify (PASS or explicit user waive).

Quality

74%

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

Fix and improve this skill with Tessl

tessl review fix ./plugin-variants/gbrain-daily/skills/meeting-ingestion/SKILL.md

The canonical home for this skill is meeting-ingestion in garrytan/gbrain

SKILL.md
Quality
Evals
Security

Quality

Content

73%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.

An exceptionally actionable, well-sequenced pipeline with strong validation gates and feedback loops — the workflow engineering is best-in-class. Its weaknesses are structural: a monolithic ~550-line file with no bundle files, inlined reference material, and noticeable repetition of the same completion guarantees.

Suggestions

Split the gbrain background mechanics (the Phase 5 facts_backstop/chronicle paragraphs and the Phase 7 auto-link/attended-edge note) into a reference file such as references/gbrain-mechanics.md, leaving a one-line pointer in each phase.

State the 'not fully ingested until verified' guarantee once in the Contract section and reference it from the HARD GATE, Output Format, and Anti-Patterns instead of restating it verbatim four times.

Verify or vendor the externally referenced files (skills/_brain-filing-rules.md, skills/conventions/*.md, docs/guides/life-chronicle.md) into the skill bundle so the progressive-disclosure structure is real and navigable.

DimensionReasoningScore

Conciseness

Almost everything is skill-specific gbrain mechanics Claude cannot be assumed to know, but the same completion guarantee is restated in at least four places (Contract, HARD GATE intro, Output Format, Anti-Patterns) and the Phase 5/7 digressions on background extraction and auto-link receipts run long — 'mostly efficient but could be tightened', not merely minor trims.

3 / 5

Actionability

Guidance is copy-paste ready throughout: exact commands ("gbrain whois <email> --json", "gbrain timeline-add {person-slug} {date} \"Attended {meeting-title}\"", "gbrain sync --source <id> --no-pull"), a complete meeting-page template, and a claim-type table with explicit verification bars.

5 / 5

Workflow Clarity

Phases 1–9 are explicitly sequenced and closed by a HARD GATE checklist (V1–V6) with deterministic checks, an explicit feedback loop ("fix → re-check → fix, until every item passes"), and a dedicated error-recovery section (e.g. revision_conflict → re-read and merge) — the full anchor-5 pattern.

5 / 5

Progressive Disclosure

No references/, scripts/, or assets/ bundle exists, so the entire ~550-line manual lives in SKILL.md, with reference-grade material (facts_backstop receipt semantics, auto-link/attended-edge mechanics, chronicle extraction) inlined in Phases 5 and 7; the external files it does cite (skills/conventions/brain-first.md, docs/guides/life-chronicle.md, ../conventions/agent-operator-protocol.md) sit outside the bundle and cannot be verified.

2 / 5

Total

15

/

20

Passed

Description

67%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.

A highly specific, distinctive description that lays out the entire pipeline, but it reads as a contract summary rather than a trigger-oriented description: it lacks a 'Use when…' clause and omits the natural user phrasings (e.g., "meeting notes") that live only in other frontmatter fields.

Suggestions

Append a 'Use when…' sentence to the description with concrete trigger phrases, e.g. "Use when the user shares a meeting transcript, meeting notes, or a recorder export, or asks to process this meeting."

Fold one or two natural synonyms users actually say ("meeting notes", "process this meeting") into the description body so trigger matching does not depend solely on the separate triggers list.

Trim the closing contract sentence ("A meeting is NOT fully ingested until…") to a shorter completion criterion — the detail is already guaranteed by the skill body.

DimensionReasoningScore

Specificity

The description enumerates the full pipeline concretely — "normalize the source into a standard transcript record, split multi-meeting recordings, resolve speakers by evidence, create the page, pass every surprising claim through the consistency check… enrich every entity, then run the verification checklist" — matching the comprehensive multi-action anchor rather than the 'minor gaps' anchor.

5 / 5

Completeness

The 'what' is clearly and thoroughly stated (ingest transcripts into brain pages with enrichment, propagation, and verification), but the description contains no 'Use when…' clause or equivalent trigger guidance, which caps completeness at 3 per the rubric guidelines.

3 / 5

Trigger Term Quality

Natural keywords like "meeting transcripts" and "meeting recorder" are present, but common user phrasings such as "meeting notes" or "process this meeting" are missing from the description field itself (they appear only in the separate triggers/when_to_use fields, which are not the evaluation target).

3 / 5

Distinctiveness Conflict Risk

"Ingest meeting transcripts from ANY meeting recorder into brain pages with attendee enrichment, entity propagation, and timeline merge" carves out a clear niche with specific target objects, making conflict with other skills' triggers minimal.

5 / 5

Total

16

/

20

Passed

Validation

81%

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

Validation — 13 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

skill_md_line_count

SKILL.md is long (593 lines); consider splitting into references/ and linking

Warning

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

relative_links

Relative link issues: 2 suspicious

Warning

Total

13

/

16

Passed

Repository

garrytan/gbrain

Path

plugin-variants/gbrain-daily/skills/meeting-ingestion/SKILL.md

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.