setup

Add keyless GBrain memory to an existing agent, or connect a hosted brain while preserving identity and configuration

Quality

78%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./skills/setup/SKILL.md

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 strong, tightly written orchestration skill: concrete commands, an explicit phased workflow with genuine validation checkpoints, error recovery, and anti-patterns. Its main weaknesses are mild repetition of the same safety warnings across sections and heavy reliance on many external documents that are not part of the skill bundle.

Suggestions

Consolidate the repeated opt-in/reinitialization warnings into the Anti-Patterns section and reference it once, trimming near-duplicate sentences in the Contract and Phase 2.

Move the dense intent-routing and optional-capability tables' longer guidance into a local references/ file to reduce inline token load and keep the external doc links to one clearly labeled section.

Inline at least one concrete example of the receipt recovery command and the search-mode matrix relay step so the common repair path is executable without following an external document.

Dimension	Reasoning	Score
Conciseness	The body is dense and imperative with no explanations of concepts Claude already knows, but a few warnings are repeated across sections — 'requires explicit opt-in' appears in the Contract, Phase 2, and the optional-capabilities table, and 'do not reinitialize' appears three times — which could be consolidated.	4 / 5
Actionability	Concrete, executable commands with specific flags appear throughout ('gbrain engine status --json', 'bun install -g github:garrytan/gbrain#latest-stable', 'gbrain init --pglite --no-embedding', 'gbrain connect ... --install'), including a warning against the unrelated npm 'gbrain' package. Minor gaps remain where steps defer entirely to external documents (e.g., 'run the receipt's exact recovery command') without any inline command.	4 / 5
Workflow Clarity	Four explicit phases give a clear sequence, and Phase 4 is a full validation loop (randomized save → readback → correction → withdrawal → verified omission → cross-conversation check), complemented by an error-recovery table with feedback loops and an anti-patterns checklist. This matches the top anchor for complex processes.	5 / 5
Progressive Disclosure	References are well-signaled, one level deep, and the routing/decision content appropriately stays in SKILL.md. However, no bundle files exist (no references/, scripts/, or assets/), all ~15 referenced paths point outside the skill directory into an external repo, and many links are embedded in dense tables, which slightly hampers navigation and prevents verifying the targets exist.	4 / 5
	Total	17 / 20 Passed

Description

75%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 specific, correctly-voiced description with good trigger coverage from the accompanying frontmatter triggers list. Its main gap is that the 'when' guidance lives outside the description sentence and a few natural trigger variations are missing.

Suggestions

Append a 'Use when...' clause to the description itself, e.g., 'Use when the user asks to set up, add, or connect GBrain memory for an agent.'

Add a couple of natural trigger phrases without the product name (e.g., 'add memory to my agent') to the triggers list to catch users who omit 'gbrain'.

Mention the key downstream capabilities (keyword recall, instruction attachment, round-trip verification) in the description to close the specificity coverage gap.

Dimension	Reasoning	Score
Specificity	'Add keyless GBrain memory to an existing agent, or connect a hosted brain while preserving identity and configuration' lists several specific actions (add keyless memory, connect a hosted brain, preserve identity/configuration) in the imperative voice. It falls short of 5 because coverage is incomplete — retrieval, instruction attachment, and verification are unnamed — but is above 3 since more than 1-2 concrete actions are given.	4 / 5
Completeness	The 'what' is clearly stated, and 'when' is covered by the frontmatter triggers list — equivalent explicit trigger guidance — rather than a 'Use when...' clause in the description itself. That earns 4 rather than 5 because the trigger guidance is not integrated into the description and could be more explicit.	4 / 5
Trigger Term Quality	The description itself carries the key terms ('GBrain', 'memory', 'agent', 'hosted brain') and the frontmatter triggers list supplies natural phrases users would actually say ('set up gbrain', 'initialize brain', 'gbrain setup', 'install gbrain into this agent workspace', 'add gbrain to my agent'). A few natural variations such as 'add memory' without the product name are missing, keeping it below 5.	4 / 5
Distinctiveness Conflict Risk	The product name 'GBrain' gives a clear niche with distinct, product-named triggers, so conflict risk is low. It is not 5 because 'connect a hosted brain' and 'add memory to an existing agent' overlap with closely related sibling skills (mcp-access, hosted harness access, bootstrap).	4 / 5
	Total	16 / 20 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
relative_links	Relative link issues: 31 suspicious	Warning

	Total	14 / 16 Passed

Repository: garrytan/gbrain
Path: skills/setup/SKILL.md
Commit: e78f1c3

Reviewed: 5 days 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.