CtrlK
BlogDocsLog inGet started
Tessl Logo

exploring-mcp-sessions

Investigate individual PostHog MCP sessions — the sequence of tool calls a single agent made in one run, what it was trying to do, and where it went wrong. Use when the user asks "what did this MCP session do?", "show me the tool calls for session X", "what was the agent's goal?", "which sessions had errors?", "who is connecting to my MCP?", or pastes an MCP analytics sessions URL.

76

Quality

95%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Quality

Content

96%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 tight, highly actionable skill body: concrete tool calls and executable SQL, well-marked sharp edges, and clear workflow sequencing with fallbacks. Progressive disclosure is good but the skill leans on inline recipes since it ships no bundle files of its own.

Suggestions

Move the four SQL recipes into a local references/ file (e.g. references/sql-recipes.md) and link to it from 'When to drop to SQL' to further trim the body and enable one-level-deep lookup.

Consider extracting the tool table plus the 'gated behind mcp-analytics flag' note into a short reference so the SKILL.md body stays a pure overview.

DimensionReasoningScore

Conciseness

Lean throughout: it opens with a one-line definition, a compact tool table, and only the sharp edges (the date-window trap, silent order_by fallback, dead distinct_id_count field) that Claude would not already know. No padding and no explanation of basic concepts.

5 / 5

Actionability

Every workflow gives a copy-paste-ready JSON tool invocation with concrete parameters, and the SQL section provides complete, executable HogQL queries with exact column and property references — fully executable, not pseudocode.

5 / 5

Workflow Clarity

Workflows are clearly sequenced (list -> read tool calls -> summarise intent) with explicit failure-mode checkpoints: the date-window trap is called out first, a 503 has a documented fallback, and the 'when to drop to SQL' section enumerates four numbered cases with recovery guidance.

5 / 5

Progressive Disclosure

Good structure with the shared schema and cross-session recipes offloaded to a clearly signaled one-level reference (models-mcp.md) and related skills linked; the only gap is that no local references/ directory exists, so the body still carries the bulk of the SQL recipes inline rather than splitting them out.

4 / 5

Total

19

/

20

Passed

Description

95%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 strong, third-person description that crisply states what the skill does and gives a rich, natural set of 'Use when' triggers including a URL-paste case. It is specific, complete, and distinct from neighboring MCP skills.

DimensionReasoningScore

Specificity

Names the domain (MCP sessions) and several concrete actions — 'the sequence of tool calls a single agent made', 'what it was trying to do', and 'where it went wrong' — which are specific rather than generic, though they describe outcomes more than enumerated tool operations.

4 / 5

Completeness

Explicitly answers both 'what' (investigate a single session's tool-call sequence, goal, and failures) and 'when' with a concrete 'Use when...' clause listing multiple trigger phrases.

5 / 5

Trigger Term Quality

Comprehensive coverage of natural phrases users would actually say, quoted verbatim ('what did this MCP session do?', 'show me the tool calls for session X', 'which sessions had errors?', 'who is connecting to my MCP?') plus the URL-paste trigger, with strong synonym coverage.

5 / 5

Distinctiveness Conflict Risk

A clearly scoped niche — individual MCP session forensics — with trigger phrases specific enough to avoid overlap with sibling skills like tool-quality or tool-usage, and a URL-based trigger that further narrows it.

5 / 5

Total

19

/

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: 6 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.