CtrlK
BlogDocsLog inGet started
Tessl Logo

implementing-agent-modes

Guidelines to create/update a new mode for PostHog AI agent. Modes are a way to limit what tools, prompts, and prompt injections are applied and under what conditions. Achieve better results using your plan mode.

57

Quality

64%

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

Fix and improve this skill with Tessl

tessl review fix ./.agents/skills/implementing-agent-modes/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

72%

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

The body is a well-structured, lean process guide with concrete file paths and commands, but it stops short of fully executable specifics in several steps and lacks explicit validation checkpoints in its workflow.

Suggestions

Add a concrete trajectory-example template or snippet in the 'Write JTBD-like trajectory examples' section instead of only pointing to the Product analytics preset.

Insert explicit validation checkpoints into the flow (e.g., 'run the mode's tests; if failing, fix tool/scaffolding and re-run') to add a validate→fix→retry feedback loop.

For the scaffolding step, show the minimal AgentToolkit/AgentModeDefinition skeleton so the guidance is copy-paste ready rather than purely descriptive.

DimensionReasoningScore

Conciseness

The body is lean and concrete — specific file paths and commands earn their tokens, and the intro paragraph explains the PostHog-internal mode architecture rather than generic concepts Claude already knows.

3 / 3

Actionability

It provides concrete file paths, commands (`hogli build:schema`), and one real code snippet, but several steps are directional without executable specifics — e.g., 'Write JTBD-like trajectory examples ... Check the Product analytics preset for reference' gives no concrete example to follow.

2 / 3

Workflow Clarity

Steps are clearly sequenced via section headers (name → schema → scaffolding → tools → toolkit → examples → frontend → flag → tests), but validation checkpoints are only implicit/scattered (verify backend-first, run tests at the end) with no explicit validate→fix→retry feedback loop.

2 / 3

Progressive Disclosure

It is a single, well-organized file with clear section navigation and no nested references; no bundle files exist, so the appropriately-contained overview with well-signaled sections fits the simple-skill case.

3 / 3

Total

10

/

12

Passed

Description

57%

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 establishes a clear, specific niche (PostHog agent modes) and names the core action, but it lacks an explicit 'Use when...' trigger and leans on technical/marketing phrasing rather than natural user keywords.

Suggestions

Add an explicit trigger clause, e.g. 'Use when creating or updating a mode for the PostHog AI agent (limiting tools, prompts, or prompt injections for a product, area, or agent).'

Replace 'Achieve better results using your plan mode' with concrete natural-language trigger terms users would actually say (e.g., 'agent mode', 'tool/prompt injection for a product area').

List the concrete actions upfront ('define, scaffold, add tools, feature-flag, and test a new agent mode') to lift specificity toward listing multiple specific actions.

DimensionReasoningScore

Specificity

It names the domain and concrete actions ('create/update a new mode for PostHog AI agent'), but does not list multiple specific actions — the remainder is a definitional clause ('Modes are a way to limit what tools, prompts, and prompt injections are applied').

2 / 3

Completeness

It clearly answers 'what' (guidelines to create/update a mode) but provides no 'Use when...' trigger; 'Achieve better results using your plan mode' is not an explicit when-condition, so the when is only implied at best.

2 / 3

Trigger Term Quality

It has some relevant keywords a PostHog developer might say ('mode', 'PostHog AI agent', 'create/update a new mode'), but 'prompt injections' and the marketing-like 'Achieve better results using your plan mode' are not natural trigger phrasings, and common variations are missing.

2 / 3

Distinctiveness Conflict Risk

The 'PostHog AI agent' mode niche is highly specific to that internal system, making it unlikely to trigger for unrelated skills despite the weak explicit triggers.

3 / 3

Total

9

/

12

Passed

Validation

100%

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

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

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.