signal-detector

Always-on ambient signal capture. Fires on every inbound message to detect original thinking and entity mentions. Spawn as a cheap sub-agent in parallel, never block the main response.

Quality

71%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Critical

Do not install without reviewing

Fix and improve this skill with Tessl

tessl review fix ./skills/signal-detector/SKILL.md

Quality

Content

57%

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

The content is well-structured and reasonably actionable with concrete commands and a clear phased workflow, but it loses points for preachy padding, incomplete command syntax on page-write operations, and the absence of validation checkpoints for its mutating writes.

Suggestions

Add an explicit validation/verification checkpoint after page writes (e.g., confirm the page exists and back-links were created) to satisfy the feedback-loop expectation for mutating operations.

Provide complete executable invocations for put_page/get_page and the enrich step instead of describing them only at the tool-name level.

Trim editorializing ('Ideas are the intellectual capital. Entities are bookkeeping. Both compound over time.') to keep the skill lean.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient with concrete commands and formats, but editorializing such as 'Ideas are the intellectual capital. Entities are bookkeeping. Both compound over time.' and emphatic aphorisms ('An unlinked mention is a broken brain.') could be tightened, matching the 'mostly efficient but includes some unnecessary explanation' anchor rather than the every-token-earns-its-place anchor at 3.	2 / 3
Actionability	It gives executable commands like 'gbrain search "name"' and 'gbrain timeline-add <slug> <date> "<summary>"' plus concrete paths and the back-link format, but core write operations ('create/update brain pages', 'trigger enrich') are described at the tool-name level rather than as complete executable invocations, fitting 'some concrete guidance but incomplete' over the copy-paste-ready anchor at 3.	2 / 3
Workflow Clarity	Phases 1-3 are clearly sequenced with a numbered entity-detection branch (no/thin/rich page states), but since the skill is mutating and batch-writes pages with no explicit validation or verify-after-write checkpoint, the rubric caps workflow clarity at 2 for missing validation in destructive/batch operations.	2 / 3
Progressive Disclosure	With no bundle files present, the body is well organized into labeled sections (Contract, Phases, Output Format, Anti-Patterns, Tools Used) and makes one clearly-signaled one-level-deep reference ('See skills/conventions/quality.md for Iron Law back-linking'), matching the clear-overview-with-well-signaled-one-level-reference anchor.	3 / 3
	Total	9 / 12 Passed

Description

85%

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 is specific, answers both what and when with an explicit always-on trigger, and occupies a distinct niche. Its main weakness is trigger-term quality: the phrasing is more technical than what a user would naturally say.

Suggestions

Add natural-language trigger phrasing (e.g., 'Use whenever the user shares an idea, observation, or mentions a person/company') so the when-clause mirrors how users actually describe needing the skill.

Soften jargon like 'ambient signal capture' and 'inbound message' in favor of user-facing terms such as 'captures ideas and mentions from your messages'.

Dimension	Reasoning	Score
Specificity	It lists multiple concrete actions — 'detect original thinking and entity mentions', 'Fires on every inbound message', and 'Spawn as a cheap sub-agent in parallel, never block the main response' — matching the 'lists multiple specific concrete actions' anchor rather than the domain-plus-some-actions anchor at 2.	3 / 3
Completeness	'Fires on every inbound message to detect original thinking and entity mentions' answers both what (detect ideas and entities) and when (every inbound message, always-on), which counts as equivalent explicit trigger guidance, so it is not capped at 2.	3 / 3
Trigger Term Quality	Phrases like 'original thinking' and 'entity mentions' are somewhat natural, but the framing ('ambient signal capture', 'inbound message') is jargon-leaning and lacks common user variations or a 'Use when...' form, so it sits at 'some relevant keywords but missing common variations' rather than full coverage at 3.	2 / 3
Distinctiveness Conflict Risk	The niche of capturing original thinking and entity mentions into brain pages is distinct from typical skills and unlikely to trigger for the wrong skill, matching the 'clear niche with distinct triggers' anchor.	3 / 3
	Total	11 / 12 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: 1 missing	Warning

	Total	14 / 16 Passed

Repository: garrytan/gbrain
Path: skills/signal-detector/SKILL.md
Commit: 3fafb69

Reviewed: about 19 hours 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.