Runs a Secret Santa name draw in chat and gives each person a private reveal link.
Scored across 3 tools
Each tool maps to a clearly distinct intent: learning the rules, choosing a gift, and configuring a group. The descriptions are explicit about what each does and does not do (e.g. the gift tool 'knows nothing about any draw'), leaving no overlap.
All three names follow the same verb_noun snake_case pattern: explain_secret_santa_rules, pick_secret_santa_gift, plan_secret_santa_setup. The domain prefix is applied uniformly and verbs are distinct and descriptive.
Three tools is on the low end but each earns its place given the narrow Secret Santa scope, and the actual draw is intentionally client-side. A slightly larger surface (e.g. a group-name or budget helper split out) isn't needed, so the count is defensible though thin.
The pre-draw lifecycle is covered: rules/fairness Q&A, gift selection by budget and interest, and setup link generation with validation. The deliberate omission of server-side draw, match retrieval, and post-draw management is a documented design boundary rather than a gap, though agents have no way to inspect or revise an existing group.
The rules and fairness answers pickmysecretsanta.com itself publishes, worded the site's way. Topics: basics (the classic rules), fairness_and_secrecy (how the sealed draw works and its honest limits), organizer_view (the organizer sees names and links, never matches), keep_apart_pairs, late_joiner (the two honest ways to add someone after the draw), budget (the four bands), white_elephant (how it differs).
| Name | Required | Description | Default |
|---|---|---|---|
| topic | Yes | Which rule to explain. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
The annotations already declare readOnly, idempotent, non-destructive, closed-world behaviour, so the burden is light. The description adds one genuinely useful trait beyond them: the answers are the site's own published wording rather than computed output ('worded the site's way'), implying static authoritative text. It does not say anything about the shape or length of what comes back.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Purpose is front-loaded in the first clause, then the topic list is compressed into one tight sentence with no filler. The parenthetical glosses are dense but each one carries information, so nothing is wasted; the slightly awkward opening clause is the only friction.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a single-parameter, read-only lookup with full schema coverage and no output schema, the description covers purpose, source, and per-topic content. The only gap is that it never says how the explanation is returned (prose block, sections) or whether topics can be combined across calls.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100%, so the baseline is 3, but the description goes further by glossing every enum value: fairness_and_secrecy covers 'how the sealed draw works and its honest limits', late_joiner the 'two honest ways', budget 'the four bands'. That tells an agent what each topic actually yields, which the bare enum names and 'Which rule to explain.' do not.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description makes clear this retrieves the site's own published rules and fairness answers, worded as the site words them, which is a specific enough resource to distinguish it from a gift-picking or setup-planning tool. It lacks an explicit verb and never names the siblings, so a reader infers rather than reads the contrast with plan_secret_santa_setup.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The enumerated topic list implies when to call (when the user asks about a rule in one of those areas), and the 'late_joiner'/'white_elephant' entries hint at specific question shapes. However, there is no explicit when-to-use statement, no exclusion, and no pointer to the sibling tools for setup or gift selection.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Filters Santa's vetted 212 hand-checked gift ideas - the Secret Santa subset of the shared 255-idea bank it shares with its sister site The Present Picker, sorted here for Secret Santa by the budget band your group agreed for its draw - with strict parameters, and returns up to three picks. Parameters: budget_band (required: under_25, 25_to_50, 50_to_100, over_100); recipient_type (anyone, kid - only ideas from the bank's Kid shelf, or adult - everything except the Kid shelf); one interest hint (books, cooking, cozy, fitness, games, gardening, music, outdoors, science, tech, travel, or surprise) which ranks matches first; and avoid, a list of words - any idea whose name, blurb, or why mentions an avoid word is excluded, and the answer says how many ideas the filters set aside. Each pick names the kind of gift, why it fits, and what to look for and avoid. No brands, no stores, no exact prices: budget bands are guides, not exact prices. This tool never asks who the gift is for by name and knows nothing about any draw.
| Name | Required | Description | Default |
|---|---|---|---|
| avoid | No | Words to exclude: any idea whose name, blurb, or why mentions one of these words is filtered out (for example "coffee" or "scented"). | |
| budget | No | Legacy alias for budget_band. | |
| interest | No | One interest hint for the person, or surprise for any. | |
| budget_band | No | The group's budget band. | |
| recipient_type | No | Who the gift is for: kid returns only the bank's Kid-shelf ideas; adult excludes them; anyone (default) applies no recipient filter. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint, idempotentHint, destructiveHint=false and openWorldHint=false, so the safety profile is covered. The description adds real behavioral detail beyond that: results are capped at three, the response reports how many ideas the filters set aside, budget bands are guides rather than exact prices, and no brands, stores, or prices appear. That is meaningful context an agent cannot get from the annotations.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Front-loads the core action, but the opening sentence is an overstuffed parenthetical about 212 versus 255 ideas and the sister site The Present Picker, which is not operationally useful. The parameter enumeration is dense but scannable. Trimming the catalog-marketing details would make the definition materially tighter.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given no output schema, the description does well to describe the shape of each pick and the filtered-out count. But for a five-parameter tool it leaves the legacy 'budget' alias undocumented and misstates which parameter is required, which are the main gaps a calling agent would hit.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100%, so the baseline is 3, and the description does add value by explaining that the interest enum ranks matches first and that avoid matches against name, blurb, or why rather than title only. However, it omits the 'budget' legacy alias entirely and declares budget_band 'required' while the schema lists no required properties, which can mislead an agent about how to call the tool.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
States a specific verb and resource: filters a curated Secret Santa gift bank and returns up to three picks, each with a kind, a why, and what to look for. That is clearly distinguishable from the sibling tools (explain_secret_santa_rules, plan_secret_santa_setup), though the description never names or contrasts them explicitly. The core action is buried after a long parenthetical about the sister site and idea counts, slightly diluting clarity.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Usage is implied rather than stated: it is the gift-selection step after a budget band has been agreed. There is no explicit when-to-use versus the setup or rules-explainer siblings, and no stated prerequisite (e.g., run plan_secret_santa_setup first). The closing line 'never asks who the gift is for by name and knows nothing about any draw' usefully bounds what the agent should not rely on it for.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Configures a Secret Santa group of 3 to 30 right here in chat and hands the organizer one pre-filled link to pickmysecretsanta.com, where the zero-knowledge draw runs in their browser - no server sees the matches, and each match will live inside one private link - hidden from plain sight, not locked. Give it the group's names and, optionally, a group name, a budget band, and keep_apart pairs (two-name lists, like [["Ana", "Ben"]], for people who must not draw each other): it validates the setup (3 to 30 people, no shared names - the site's draw cannot tell two same-named people apart) and checks that a valid draw exists under the keep-apart pairs before it returns the link with the group, names, budget, and pairs already in the form, plus the site's honest warnings; if the pairs leave no fair draw, it names the conflict plainly and returns no link. The link carries setup only - nobody has been assigned anyone. Names are used once to build the link and are never stored. With only a headcount (group_size) and no names, it instead returns the setup checklist the site prescribes: what to agree out loud first (buy-by date, what counts as a gift, the late-joiner rule, the reveal), the budget step, and how the browser draw works.
| Name | Required | Description | Default |
|---|---|---|---|
| names | No | Every player's name, one per person (3 to 30). Names must be distinct - add a last initial where two people share a name. When names are given, the tool returns the pre-filled setup link. | |
| budget | No | Legacy alias for budget_band. | |
| group_name | No | Optional group name, up to 40 characters (for example "The Office Crew"). | |
| group_size | No | How many people, 3 to 30 - for the checklist when no names are given. | |
| keep_apart | No | Pairs who must not draw each other, as two-name lists using names from the names list, e.g. [["Ana", "Ben"]]. | |
| budget_band | No | The group's budget band, if already picked. | |
| has_keep_apart_pairs | No | Checklist mode only: true if some pairs must not draw each other. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
The description discloses substantial behavior beyond the annotations: the zero-knowledge browser draw, that no server sees matches, that names are never stored, that the returned link carries setup only without assignments, and that invalid keep-apart configurations return no link with the conflict named. This richly supplements the readOnlyHint/idempotentHint annotations with privacy, validation, and failure-mode context.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is front-loaded with the core purpose and contains no obviously wasted sentences, but it is a dense, run-on paragraph. The second sentence is especially long with nested parentheticals, which slightly hurts readability even though every clause conveys relevant information.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a read-only, two-mode setup tool with seven parameters and no output schema, the description covers what the tool does, how each mode behaves, what is returned in success and failure, and the privacy model. An agent has everything needed to invoke it correctly.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
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 adds useful cross-parameter semantics: it explains that names and group_size select different modes, clarifies that keep_apart entries are two-name lists referencing names from the names list, and reinforces that names must be distinct. These details go beyond the per-parameter schema text.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description states a specific verb and resource: it configures a Secret Santa group and returns a pre-filled setup link to pickmysecretsanta.com. It clearly distinguishes the two operating modes (names → link, headcount → checklist). It does not explicitly differentiate itself from the sibling tools explain_secret_santa_rules and pick_secret_santa_gift, so it falls short of a 5.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description makes the two usage modes clear: provide names to get a pre-filled link, or provide only group_size to get the site's setup checklist. However, it offers no explicit guidance on when to use this tool versus the sibling tools, nor does it state any exclusions or prerequisites beyond the validation rules.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
explain_secret_santa_rulespick_secret_santa_giftplan_secret_santa_setupDraw names for Secret Santa without sign-ups or email addresses.
Organise a Secret Santa from Claude or ChatGPT. Matchings never appear in the chat.
Manage gift wishlists, reserve gifts on friends' lists and run a Secret Santa from a chat.
Fair random picks with the wheel in your chat: pick names, split teams, run certified draws.
Add one secure layer between your agents and this server.