Skip to main content
Glama
Hermoso

Hermoso

Official

Which hooks and subjects are getting traction

post_performance
Read-onlyIdempotent

Groups a brand's published posts by hook, subject, format, channel, media, or hour to report engagement rates, rank top and bottom posts per channel, and suppress verdicts from fewer than five posts.

Instructions

Use this when the user asks which of their posts, hooks or subjects work best. It groups this brand's published posts by hook (default), subject, format (recipe), channel, media format or posting hour, reports the engagement RATE within each channel, and ranks the best and worst POSTS in each channel. Describe a post by the creative it carried (what it shows, its format, its link), not by its caption — the caption is the least important part of a post. THREE THINGS IT WILL NOT DO, to repeat rather than paper over: (1) it never sums metrics across channels — a LinkedIn impression and a TikTok view are different units, so every comparison is within one channel; (2) it SUPPRESSES a verdict below 5 measured posts and says so, because a confident recommendation from 3 posts is worse than none; (3) a post with no recorded hook (published outside Hermoso, or backfilled without a creation match) counts toward channel and format totals but never votes on which hook works. Present the finding verbatim if there is one, and the reason if there is not. Read-only, 0 credits.

Tool call template

This tool declares no required parameters. Add optional parameters from the input schema as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {},
    "name": "post_performance"
  }
}

Input Schema

NameRequiredDescriptionDefault
axisNowhat to group by — default hook; recipe = the format of the creative
daysNolook back N days (1-730) over the whole history; omit for the recent posts only
brandNoWHICH PROFILE the post lives in — id or exact name from list_brands (a shared workspace: its profile id). Needed when it was published in a profile this connection is not pinned to; applies to THIS CALL ONLY. A name that matches no profile, or two, is REFUSED and nothing is done.
channelNorestrict to one channel

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changedv0.1.374
    • changedInput schema / properties / brand / description
      Previous value: -"WHICH BRAND the post lives in — id or exact name from list_brands (a shared workspace: its profile id). Needed when it was published in a brand this connection is not pinned to; applies to THIS CALL ONLY. A name that matches no brand, or two, is REFUSED and nothing is done."New value: +"WHICH PROFILE the post lives in — id or exact name from list_brands (a shared workspace: its profile id). Needed when it was published in a profile this connection is not pinned to; applies to THIS CALL ONLY. A name that matches no profile, or two, is REFUSED and nothing is done."
  2. Changed1 schema field changedv0.1.320
    • changedInput schema / properties / brand / description
      Previous value: -"WHICH BRAND the post lives in — the id or exact name from list_brands (a workspace shared with you: its profile id). Needed when it was published in a brand this connection is not pinned to: a post you can CREATE in a brand is manageable there too, for THIS CALL ONLY, without switching the connection. A name that matches no brand, or two, is REFUSED and nothing is done."New value: +"WHICH BRAND the post lives in — id or exact name from list_brands (a shared workspace: its profile id). Needed when it was published in a brand this connection is not pinned to; applies to THIS CALL ONLY. A name that matches no brand, or two, is REFUSED and nothing is done."
  3. Changed1 schema field changedv0.1.281
    • addedInput schema / properties / days
      Added value: +{
      +  "description": "look back N days (1-730) over the whole history; omit for the recent posts only",
      +  "type": "number"
      +}
  4. Changed1 schema field changedv0.1.272
    • addedInput schema / properties / brand
      Added value: +{
      +  "description": "WHICH BRAND the post lives in — the id or exact name from list_brands (a workspace shared with you: its profile id). Needed when it was published in a brand this connection is not pinned to: a post you can CREATE in a brand is manageable there too, for THIS CALL ONLY, without switching the connection. A name that matches no brand, or two, is REFUSED and nothing is done.",
      +  "type": "string"
      +}
  5. Changed2 schema fields changedv0.1.227
    • changedInput schema / properties / axis / description
      Previous value: -"what to group by — default hook"New value: +"what to group by — default hook; recipe = the format of the creative"
    • changedInput schema / properties / axis / enum
      Previous value: -[
      -  "hook",
      -  "subject",
      -  "channel",
      -  "media",
      -  "hour"
      -]New value: +[
      +  "hook",
      +  "subject",
      +  "recipe",
      +  "channel",
      +  "media",
      +  "hour"
      +]
  6. Addedv0.1.161

TDQS

A4.4/5.0
Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare read-only/idempotent/non-destructive, and the description still adds substantial behavior the agent could not infer: comparisons are always within a single channel, verdicts are suppressed below 5 measured posts, hookless posts count toward totals but never vote, the `finding` must be presented verbatim, and the call costs 0 credits.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The trigger sentence and the grouping/ranking behavior are front-loaded where they belong, and the numbered constraints are easy to scan. Some editorializing ('to repeat rather than paper over', em-dash asides, a parenthetical about backfilled posts) could be trimmed without losing meaning.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a read-only analytics tool with no output schema, the definition covers what is grouped, what unit is compared, when no verdict appears, how to present the `finding`, and the cost. An agent has everything needed to call it and to report results without overclaiming.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100% and the axis enum already documents 'recipe = the format of the creative', so the description largely restates it. The one added value — describe a post by its creative rather than its caption — is about interpreting output, not about parameter meaning. Baseline 3 is appropriate.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a concrete verb set — group published posts by hook/subject/format/channel/media/hour, compute within-channel engagement rate, rank best and worst posts — plus the exact user question it answers. This clearly separates it from list_hooks (enumeration) and collect_post_metrics (raw metrics) without needing to open the schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

'Use this when the user asks which of their posts, hooks or subjects work best' is an explicit trigger condition. The three 'will not do' clauses set scope boundaries (no cross-channel summation, no verdict under 5 posts, no hook voting for untagged posts), but no sibling tool is named as an alternative for those excluded cases.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Deploy Server

Other Tools