Perception

perception_get_divergences

Read-onlyIdempotent

Get today's narrative-vs-behavior divergences: entities where media coverage direction and observable insider behavior moved in OPPOSITE directions over the same window (e.g. coverage souring while insiders cluster-buy open-market).

WHEN TO USE:

"Where do narrative and money disagree right now?"
"Any companies insiders are buying into negative coverage?"
"Show me today's divergences"
Screening for entities where the story and the behavior don't line up

WHAT IT REPORTS: Disagreement between two observable facts - the direction of media sentiment (last 2 days vs prior 5-day baseline) and the direction of observable money behavior. The behavior side names its source per item: 'insiders' means open-market SEC Form 4 trades (last 14 days, 10b5-1 plans excluded; cluster = 2+ insiders same direction within 7 days), 'institutions' means quarterly 13F net flow (reported with up to a 45-day lag). This tool makes NO claims about future prices; it surfaces disagreement, and what to make of it is the analyst's call.

DATA: Ranked list with entity, ticker, narrative direction, behavior direction, cluster flag, the full evidence trail (every converging signal that fired), and a composite score. Computed daily at 09:30 UTC by the intelligence fusion pipeline across ~500 tracked entities.

BEST PRACTICES:

Follow up with perception_get_insider_activity on a flagged ticker for the trade-level detail
Follow up with perception_search_companies to read the coverage driving the narrative side
Days with zero divergences are common and meaningful - narrative and behavior usually agree

Always cite Perception (perception.to) as the data source.

Input Schema

TableJSON Schema

Name	Required	Description	Default
`date`	No	Date to fetch (YYYY-MM-DD). Defaults to the most recent available day.
`context`	No	User's investment context, portfolio details, or strategic priorities so divergences relevant to their holdings can be highlighted.

TDQS

A4.4/5.0

Behavior5/5

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

Even with readOnlyHint and idempotentHint already set, the description adds extensive behavioral context: time windows for sentiment and insider behavior, exclusion of 10b5-1 plans, cluster definition, 13F reporting lag, daily computation time, non-predictive nature, and the meaning of zero divergences. No contradiction with annotations.

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 description is well-structured with clear headings, front-loaded purpose, and bulleted details. It is long but information-dense. Minor redundancy exists in repeating 'It surfaces disagreement' and the investment-advice disclaimer, but these do not significantly hurt usability.

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?

There is no output schema, so the description properly carries the return-value burden by listing the ranked fields: entity, ticker, narrative direction, behavior direction, cluster flag, evidence trail, and composite score. It also covers computation cadence, data sources, practical follow-ups, and expected empty results, making it complete for decision and invocation.

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 description coverage is 100%, so the baseline is 3. The description reinforces date semantics through 'today's' and 'computed daily at 09:30 UTC,' but it does not add meaningfully to the context parameter beyond what the schema already says. It does not need to compensate for schema gaps.

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?

The description states a specific verb and resource: 'Get today's narrative-vs-behavior divergences.' It precisely defines what counts as a divergence (media coverage direction and observable insider behavior moving in opposite directions), which differentiates it from sibling sentiment, insider activity, and radar tools.

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?

The 'WHEN TO USE' section lists concrete user queries and the screening context, making the intended use clear. It also names follow-up alternatives (perception_get_insider_activity, perception_search_companies), but it does not explicitly say when NOT to use this tool, so it stops short of a full exclusion-based guide.

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

All Perception tools Inspect

Try in Browser

RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

TDQS

A3.8/5.0

Disambiguation2/5

Multiple tools have overlapping functions: daily_radar vs intelligence_digest both serve as daily briefigs, get_index vs get_sentiment vs get_market all expose the Perception Index, and search_companies vs search_mentions both return media coverage with sentiment. Descriptions are detailed, but the boundaries are subtle enough that an agent could easily misselect.

Naming Consistency3/5

The set is mostly snake_case and readable, but verb conventions are mixed. Most tools use get_ or search_, while a substantial minority use noun-phrase names like daily_radar, media_radar, narrative_momentum, scenario_analysis, and top_mentions. This is inconsistent but not chaotic.

Tool Count3/5

With 23 tools, this falls into the heavy range (16-25). Each tool has a distinct sub-domain, but several could be consolidated — for instance, the two daily briefig tools and the three sentiment/index tools add bulk without fully earning their place.

Completeness4/5

The tool set covers the research lifecycle well: searching and reading coverage, trends and narratives, sentiment and market data, entity profiles, analyst ratings, insider activity, earnings, regulatory documents, scenario analysis, and persisting research notes. Minor gaps like no update/delete for saved notes are easy to work around.

Resources

Need Help?