Name: signals-scout-replay-vision
Rating: 63.2 (1 reviews)
Author: PostHog

signals-scout-replay-vision

Signals scout for PostHog Replay Vision scanners. Watches that enabled scanners keep observing (throughput / quota cliffs) and that what they see in aggregate gets surfaced (score shifts, recurring themes across sessions), and files each validated finding as a report in the inbox.

Quality

76%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./products/signals/skills/signals-scout-replay-vision/SKILL.md

Quality

Content

77%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 highly actionable, well-sequenced scouting skill with concrete SQL, named tools, and strong validation gates; its main weaknesses are noticeable cross-section repetition and a monolithic body that does not split its dense SQL patterns and footguns into bundled reference files.

Suggestions

De-duplicate the emits_signals push/pull dedupe guidance into one section and cross-reference it, removing the repeated prose from the MCP-tools and disqualifier sections.

Move the five 'Vision SQL footguns' and the example SQL queries into a references/vision-sql.md file, keeping only the roster/pulse anchor query and the decision table inline.

Consolidate the 'scanner edits aren't in the activity log' note to a single stated-once reminder instead of repeating it in the orientation, watch-gap, and MCP-tools sections.

Dimension	Reasoning	Score
Conciseness	Mostly efficient domain-specific content with no generic 'what is a scanner' filler, but it repeats several points across sections (emits_signals push/pull dedupe courtesy appears in three places; 'scanner edits aren't in the activity log' recurs multiple times) that could be tightened.	3 / 5
Actionability	Fully executable guidance: multiple copy-ready SQL queries with exact properties and bounds, named MCP tools with parameters, concrete thresholds (≥30 sessions/week, <~40% drop), and exact scratchpad key formats covering the common cases.	5 / 5
Workflow Clarity	Clear run sequence (Get oriented → Explore → Save memory → Decide → Close out) with explicit validation gates — quick close-out posture check, inbox dedupe-before-author, edit-vs-author rules, 'When to stop' conditions, and the 'confirm push path is live / require distinct-session spread before authoring' feedback loops.	5 / 5
Progressive Disclosure	Well-organized with clear headers and it correctly offloads the generic report contract to a signaled one-level reference (authoring-scouts → references/report-contract.md), but the body itself is a ~265-line monolith with the SQL footguns and pattern queries inlined rather than split into reference files, and the skill's own bundle has no references/ files.	3 / 5
	Total	16 / 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, well-scoped description that clearly distinguishes the skill and names concrete capabilities, but it omits an explicit 'Use when...' trigger clause, which caps its completeness. Adding a natural trigger sentence would lift the completeness and trigger-term scores.

Suggestions

Add an explicit trigger clause, e.g. 'Use when triaging PostHog Replay Vision scanner health or when aggregate scanner output (verdict rates, scores, tag/theme concentration) shifts across sessions.'

Soften jargon with one or two plain-language synonyms a non-specialist user might say (e.g. 'session recording scanners' alongside 'Replay Vision scanners').

State the report outcome up front in plainer terms ('files a cross-session finding as an inbox report') so the 'when' is as concrete as the 'what'.

Dimension	Reasoning	Score
Specificity	Names multiple concrete actions — 'Watches that enabled scanners keep observing (throughput / quota cliffs)', 'score shifts, recurring themes across sessions', 'files each validated finding as a report in the inbox' — covering both observing-integrity and aggregate-surfacing plus report filing comprehensively.	5 / 5
Completeness	It clearly answers 'what' but offers no 'Use when...' clause or equivalent explicit trigger guidance; per the rubric a missing trigger clause caps completeness at 3.	3 / 5
Trigger Term Quality	Domain-accurate keywords a Replay Vision operator would say ('PostHog Replay Vision scanners', 'throughput', 'quota cliffs', 'score shifts', 'recurring themes') give good coverage, but the language is jargon-heavy and a few more natural surface-level phrasings are absent.	4 / 5
Distinctiveness Conflict Risk	'Signals scout for PostHog Replay Vision scanners' carves a clear niche with distinct triggers and minimal overlap with sibling scouts (session-replay, error-tracking) which it explicitly disclaims.	5 / 5
	Total	17 / 20 Passed

Validation

81%

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 — 13 / 16 Passed

Validation for skill structure

Criteria	Description	Result
metadata_version	'metadata.version' is missing	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
referenced_paths_exist	Referenced path issues: 3 missing	Warning

	Total	13 / 16 Passed

Repository: PostHog/posthog
Path: products/signals/skills/signals-scout-replay-vision/SKILL.md
Commit: 31f2a90

Reviewed: about 23 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.