CtrlK
BlogDocsLog inGet started
Tessl Logo

exploring-mcp-intent-clusters

Explore PostHog MCP intent clusters — agent goals grouped by semantic similarity, with each cluster's tool distribution and error rates, plus the tool-centric pivot (capture rate per intent, discovery rate against the advertised catalog, description fit, tool overlaps). Use when the user asks "what are agents trying to do with the MCP?", "group the intents", "which goals fail most?", "what does each cluster route to?", "when agents have this intent do they find my tool?", "which tools get mixed up?", wants to recompute the clustering, or pastes an MCP analytics intent-clustering URL.

71

Quality

87%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

75%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 well-structured, information-dense skill body that gives Claude the exact field semantics and gotchas needed to use two typed tools correctly, with concrete invocations and sensible workflow sequencing. It is consistently strong rather than exceptional on any single dimension.

Suggestions

Show the recompute poll loop and the cluster-entry join as small code/JSON snippets instead of prose so the guidance is fully copy-paste ready.

Tighten the session-summary exclusion paragraph to the essential rule and the coverage check, trimming the rationale.

Consider extracting the per-field reference (cluster and tool-pivot field lists) into a references file so SKILL.md reads as a leaner overview.

DimensionReasoningScore

Conciseness

The prose is dense and almost every section adds non-obvious, domain-specific gotchas (caps, coverage bounds, attribution rules) rather than generic padding; a few passages such as the session-summary exclusion rationale could be trimmed slightly.

4 / 5

Actionability

Provides two copy-paste-ready JSON tool invocations, exact field names to read, an exact UI link URL, and a concrete SQL tip; minor gaps are that the join logic and recompute poll loop are described in prose rather than shown as runnable examples.

4 / 5

Workflow Clarity

Distinct labeled workflows (read clusters, discoverability, empty/stale handling, recompute) are clearly sequenced with an explicit poll-until-idle/error feedback loop and a 'read coverage before quoting numbers' checkpoint; only minor validation guidance is implicit.

4 / 5

Progressive Disclosure

Well-organized with clear section headers and a properly signaled 'Related skills' section linking one level out to sibling skills; no bundle files exist and the single-file overview is appropriate, though it is a fairly long monolithic body.

4 / 5

Total

16

/

20

Passed

Description

100%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.

An exemplary description: concrete capabilities, a thorough set of natural trigger phrases, and an explicit 'Use when' clause, all in third-person voice and tightly scoped to a distinct niche. No vague fluff or over-claims.

DimensionReasoningScore

Specificity

Lists multiple concrete capabilities — 'tool distribution and error rates', 'capture rate per intent', 'discovery rate against the advertised catalog', 'description fit', 'tool overlaps', and recompute — giving comprehensive coverage of what the skill surfaces.

5 / 5

Completeness

Clearly answers 'what' (the cluster snapshot and tool-centric pivot) and explicitly answers 'when' via 'Use when the user asks...' with concrete trigger phrases, satisfying both halves.

5 / 5

Trigger Term Quality

Includes a rich set of natural quoted phrases a user would actually say ('what are agents trying to do with the MCP?', 'group the intents', 'which goals fail most?', 'which tools get mixed up?') plus a URL-paste trigger, covering synonyms and variations comprehensively.

5 / 5

Distinctiveness Conflict Risk

Occupies a clear niche (PostHog MCP intent-clustering analytics) with triggers specific enough that it would not fire for sibling analytics skills; minimal conflict risk.

5 / 5

Total

20

/

20

Passed

Validation

93%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

relative_links

Relative link issues: 2 suspicious

Warning

Total

15

/

16

Passed

Repository
PostHog/posthog
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.