CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

maintain

Brain health checks: back-link enforcement, citation audit, filing validation, stale info detection, orphan pages, and benchmarks. Use when asked to check brain health, run maintenance, or audit quality.

Quality

75%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Medium

Suggest reviewing before use

Fix and improve this skill with Tessl

tessl review fix ./plugin-variants/gbrain-coding/skills/maintain/SKILL.md

The canonical home for this skill is maintain in garrytan/gbrain

SKILL.md
Quality
Evals
Security

Quality

Content

60%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 delivers exceptionally concrete, executable gbrain guidance with genuine validation checkpoints and consent gating for paid/destructive operations, but it is a 31KB monolith. Deep system-internal detail (dream-cycle config knobs, migration repair, heartbeat ops) is inlined instead of split into reference files, and version-numbered notes are scattered rather than isolated, hurting token efficiency and navigation.

Suggestions

Split the dream-cycle section (~150 lines of triage/rescue/cooldown/quote-verify detail) into a references/dream-cycle.md file, keeping only the invocation patterns and the Iron Law quality bar in SKILL.md.

Move the Heartbeat Integration and 'Fix a half-migrated install' sections into separate reference files and keep one-line pointers in the main body, cutting the main file to a scannable overview.

State the embedding-consent procedure once (e.g., in an 'Embedding freshness' section) and reference it from Weekly maintenance instead of repeating the commands, and consolidate version-specific notes (v0.36.4.0, v0.23, v0.11.x) into a single versioned changelog/deprecations reference.

DimensionReasoningScore

Conciseness

At 567 lines the body inlines ~150 lines of dream-cycle internals (triage gates, rescue knobs, cooldown semantics, issue IDs #6377/#6753) that read as system documentation rather than maintenance instructions, repeats the embedding-consent procedure verbatim in two sections, and scatters time-sensitive version numbers (v0.36.4.0, v0.23, v0.11.1+) outside any deprecated section — noticeably verbose with several padded sections.

2 / 5

Actionability

Guidance is fully executable throughout: exact commands with flags and placeholders ("gbrain doctor --remediate --yes --include-repairs --expect <plan_hash> --max-usd 5"), concrete exit-code semantics (0/1/2 for autopilot, 3 for consent, 11 for budget, 17/18 for alerts), specific config keys, and copy-paste bash snippets covering the common cases.

5 / 5

Workflow Clarity

The manual path is numbered with real checkpoints (health before/after, preview→consent→apply for paid repairs, exit-code gating, dry-run before embed) and a 'When it fails' recovery section, matching the clear-sequence anchor. It is not a 5 because the hierarchy is muddled: step '2. Check each dimension' is followed by dozens of same-level '###' sections never numbered as steps, and the periodic workflows (Benchmark Testing, Heartbeat Integration) sit outside any explicit sequence.

4 / 5

Progressive Disclosure

No bundle files exist at all (no references/, scripts/, or assets/), and the body inlines large blocks that clearly belong in separate files — the dream-cycle configuration reference, the heartbeat integration procedures, and the half-migrated-install repair — while the few doc links point at repo-relative paths rather than organized skill references. This matches the 'content that clearly belongs in separate files is inlined' anchor.

2 / 5

Total

13

/

20

Passed

Description

83%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 strong description: third-person, concise, with a concrete capability list and an explicit 'Use when' clause carrying natural trigger phrases. Its main weaknesses are coverage gaps (the dream/synthesize workflow is a major skill capability yet unmentioned) and a couple of generic trigger terms that dilute distinctiveness.

DimensionReasoningScore

Specificity

The description lists six named concrete actions ("back-link enforcement, citation audit, filing validation, stale info detection, orphan pages, and benchmarks"), matching the several-specific-actions anchor, but it omits the dream/synthesize cycle that the body devotes ~150 lines to — a capability gap beyond 'minor', so not a 5.

4 / 5

Completeness

It explicitly answers both questions: 'what' via the enumerated checks and 'when' via "Use when asked to check brain health, run maintenance, or audit quality" — concrete trigger phrases attached to a clear capability list, matching the top anchor.

5 / 5

Trigger Term Quality

"check brain health", "run maintenance", and "audit quality" are natural phrases a user would say, giving good keyword coverage; however common variations the frontmatter triggers list itself contains ("orphan pages", "stale pages", "check backlinks", "run dream") are absent from the description, so it falls short of comprehensive.

4 / 5

Distinctiveness Conflict Risk

"Brain health checks" plus "back-link", "orphan pages", and "citation audit" carve a distinct niche, but the 'when' triggers "run maintenance" and "audit quality" are generic and could overlap with unrelated maintenance/audit skills, leaving minor overlap risk rather than minimal.

4 / 5

Total

17

/

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 (610 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: 3 missing, 1 suspicious

Warning

Total

13

/

16

Passed

Repository

garrytan/gbrain

Path

plugin-variants/gbrain-coding/skills/maintain/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.