Name: working-outside-the-console
Rating: 66.4 (1 reviews)
Author: PostHog

working-outside-the-console

Operating without a UI — MCP / IDE / Slack mode. How to compensate for missing client tools, how to be useful in a text-only chat. Load when the session client kind is NOT `posthog-code`.

Quality

80%

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 ./products/agent_platform/backend/kernel_skills/working-outside-the-console/SKILL.md

Quality

Content

85%Weight 40%Scale 1-3

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

The body is well-structured and actionable with concrete templates and clear decision logic, scoring well on actionability, workflow clarity, and progressive disclosure. Its main weakness is conciseness: a speculative future-mode section and some re-stated detection content consume tokens that could be trimmed.

Suggestions

Cut or condense the 'The slack mode (when it exists)' section to a one-line forward note, since it cannot fire in the current version and adds ~13 lines of non-actionable content.

Tighten 'Detecting that you're outside PostHog Desktop' to avoid restating the client-kind list already given in the intro; lead with the tool-surface signal it identifies as the reliable check.

Consider collapsing the comparison table's redundant prose into the table itself to reduce surrounding explanation.

Dimension	Reasoning	Score
Conciseness	The body is mostly lean and skill-specific without over-explaining known concepts, but the speculative 'The slack mode (when it exists)' section (~13 lines that will not fire in v0) and the 'Detecting that you're outside PostHog Desktop' section restating the intro add tokens that could be tightened.	2 / 3
Actionability	For an instruction-only skill it gives concrete, copy-pasteable templates (the blockquote reference/summary examples) and specific numeric guidance ('1-3 tool calls per turn', 'A 200-line agent.md is OK; a 2000-line source is not'), which is actionable rather than abstract.	3 / 3
Workflow Clarity	Compensating moves are sequenced (1-3) and the 'Detecting' section lays out explicit decision branches (focus_*/toast present → PostHog Desktop; absent → text-only); no destructive or batch operations require validation loops here.	3 / 3
Progressive Disclosure	A single self-contained SKILL.md organized into well-labeled sections with no need for external bundle references (no references/scripts/assets exist); the inline file paths are illustrative user-facing examples, not nested skill references, so well-organized sections suffice for a 3.	3 / 3
	Total	11 / 12 Passed

Description

75%Weight 40%Scale 1-3

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 clearly answers both what and when with a distinctive trigger, but its capability list is high-level and its trigger phrasing leans on technical jargon rather than natural user language. Strong on completeness and distinctiveness, weaker on specificity and trigger-term naturalness.

Suggestions

Replace or supplement the jargon trigger 'session client kind is NOT posthog-code' with natural user phrasings like 'Use when connected via MCP, an IDE, or Slack, or when no UI is available'.

Expand the capability statement into concrete actions, e.g. 'Summarize artifacts in text, inline references and slugs, paste code blocks, and sequence tool calls for UI-less sessions'.

Drop 'Slack mode' from the description if it is not yet supported, to avoid implying a capability the skill does not currently cover.

Dimension	Reasoning	Score
Specificity	Names the domain ('Operating without a UI — MCP / IDE / Slack mode') and two actions ('compensate for missing client tools', 'be useful in a text-only chat'), but these are high-level rather than a comprehensive list of concrete actions, matching the score-2 anchor rather than the multi-action score-3 anchor.	2 / 3
Completeness	Explicitly states what it does ('How to compensate for missing client tools, how to be useful in a text-only chat') and when to load it ('Load when the session client kind is NOT posthog-code'), clearly answering both what and when with an explicit trigger.	3 / 3
Trigger Term Quality	Includes relevant keywords (MCP, IDE, Slack, text-only chat) but the explicit trigger 'session client kind is NOT posthog-code' is technical jargon a user would not naturally say, and common phrasings are missing — some relevant keywords but not full natural coverage.	2 / 3
Distinctiveness Conflict Risk	It carves a clear niche (UI-less MCP/IDE/Slack operation) with a distinctive negative trigger tied to client kind, making it unlikely to fire for unrelated skills.	3 / 3
	Total	10 / 12 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: PostHog/posthog
Path: products/agent_platform/backend/kernel_skills/working-outside-the-console/SKILL.md
Commit: 047ca00

Reviewed: 18 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.