CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

media-ingest

Ingest video, audio, PDF, book, screenshot, and GitHub repo content into the brain. Multi-format handling with entity extraction and backlink propagation. Covers video-ingest, youtube-ingest, and book-ingest subtypes.

Quality

77%

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 ./skills/media-ingest/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

81%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.

A well-structured, actionable skill body: the five-phase workflow is explicitly sequenced and validated, commands and templates are concrete, and pitfalls/anti-patterns are specific rather than generic. Remaining gaps are modest — some duplicated error-handling guidance, slightly vague fetch instructions in Phase 1, and externally referenced files that are not part of this bundle.

DimensionReasoningScore

Conciseness

Dense, high-signal body with tables, a phase structure, and no explanations of concepts Claude already knows. Minor trim opportunities: the retry policy and '[transcript unavailable]' handling are each stated twice (Error Handling vs. When it fails), and the anti-pattern note about sanctioned synthesis output is convoluted.

4 / 5

Actionability

Concrete commands ('gbrain files upload-raw <file> --page <slug>', 'gbrain sync'), a complete page template, a numbered 4-step entity loop, and an exact output message. Falls short of fully copy-paste-ready because Phase 1 fetch actions remain high-level ('Fetch transcript (Whisper, transcription service, or captions)') with no concrete command per format.

4 / 5

Workflow Clarity

Five clearly sequenced phases with explicit validation checkpoints: duplicate detection before page creation (search by source URL/hash), the OCR <80%-readable quality gate, retry-once on transient API failures, and the 'NOT fully ingested until entity propagation is complete' completion criterion. The destructive/batch cap does not apply since validation is present.

5 / 5

Progressive Disclosure

Well-organized sections with clearly signaled, one-level-deep references ('Read skills/_brain-filing-rules.md before creating any new page', 'See skills/conventions/quality.md'). However, no bundle files exist to verify these targets, the path styles are inconsistent ('skills/...' vs '../conventions/...'), and the Error Handling / When it fails content overlaps.

4 / 5

Total

17

/

20

Passed

Description

66%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 solid, specific description that names concrete actions and input formats in third person, but it lacks an explicit 'when to use' clause, which caps completeness at 3 and keeps trigger quality and distinctiveness just below their top anchors. Folding one 'Use when...' sentence with natural user phrases into the description would lift it substantially.

Suggestions

Add an explicit 'Use when...' clause to the description, e.g. 'Use when the user shares a video, YouTube link, podcast, PDF, book, screenshot, or GitHub repo and wants it saved into the brain.'

Include natural synonyms and file extensions currently missing from the description (podcast, transcript, .pdf, .mp3) to reach comprehensive trigger coverage.

Distinguish the skill from generic media-processing skills by stating the brain-page outcome in the trigger guidance (e.g. 'Use when the user asks to ingest or save media into their brain'), reducing overlap risk with single-format skills.

DimensionReasoningScore

Specificity

Names the domain ('Ingest video, audio, PDF, book, screenshot, and GitHub repo content into the brain') and several concrete actions ('entity extraction and backlink propagation'), but coverage has minor gaps — transcription, page creation, and summarization are only implied by 'multi-format handling'.

4 / 5

Completeness

The 'what' is clear (multi-format ingest into the brain with entity extraction and backlink propagation), but the description field contains no 'Use when...' clause or equivalent explicit trigger guidance, capping completeness at 3 per the judging guidelines.

3 / 5

Trigger Term Quality

Good natural-term coverage: 'video, audio, PDF, book, screenshot, and GitHub repo' plus 'youtube-ingest' subtype mention. A few common terms are missing (podcast, .pdf/.mp3 extensions), so it falls short of the comprehensive-synonyms anchor 5.

4 / 5

Distinctiveness Conflict Risk

'Ingest ... into the brain' with entity extraction is a distinctive niche, but the broad multi-format list (PDF, video, screenshot, repo) creates minor overlap risk with single-format skills when the brain context is absent.

4 / 5

Total

15

/

20

Passed

Validation

87%

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

Validation — 14 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

frontmatter_unknown_keys

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

Warning

relative_links

Relative link issues: 1 suspicious

Warning

Total

14

/

16

Passed

Repository

garrytan/gbrain

Path

skills/media-ingest/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.