Sundae Bar Logo
Past Challenge
Manager Coaching and 1-on-1 Preparation Skill
SBC38

Manager Coaching and 1-on-1 Preparation Skill

Overview

This skill teaches the agent to prepare a structured, printable 1-on-1 brief that a manager can walk into a meeting with, built from raw signals like action-item logs, sprint figures, Slack excerpts, and draft agenda notes.

Given any available combination of those inputs, the agent produces a single brief, 350 to 700 words, containing a synthesised context header, a deliberately ordered and classified agenda, signal-tethered talking points for each item, and a parking lot for topics that belong elsewhere.

The submission is a SKILL.md, with optional reference files the agent opens by name when SKILL.md directs it. A strong skill teaches the agent a reasoning procedure: how to read signals, classify agenda items, detect reclassification triggers, route sensitive topics, and structure output, so it can handle inputs it has never seen before. A weak skill provides templates or pre-filled answers that break as soon as names, figures, or topic combinations rotate.

A skill built to memorise specific scenarios will not hold up as challenges evolve.

You submit a skill: a SKILL.md and, if you need them, reference files it points at. A fixed base agent loads it and answers all 8 scenarios with it, then returns a scored breakdown. The submission that holds the top of the leaderboard accrues on-chain emissions for as long as it stays first.

We are looking for skills that demonstrate good design and help the agent perform like a capable practitioner in this domain.

Requirements (8 Tests)

Your skill is tested across 7 capability types, with 8 scenarios in total. Every submission is evaluated against all 8 scenarios. There is no sampling, so consistent quality across every capability type counts.

CategoryOverviewTest cases
Type A: Signal-Tethered Coaching and Career QuestionsThis type tests whether the agent writes coaching and career talking points as open questions tied to a specific, named signal from the input. The skill must teach the agent to identify the most relevant signal first, then construct a question from it, and to treat generic or untethered questions as a disqualifying failure mode rather than an acceptable default.1
Type B: Agenda Item Classification and Recognition SpecificityThis type tests whether the agent assigns exactly one conversation-type tag to each agenda item based on the evidence in the input, independent of the manager's label. The skill must teach the agent to distinguish between recognition backed by a genuine signal and recognition that would have to be manufactured, and to omit the recognition opener with an explicit note when no real signal exists.1
Type C: Classification Threshold JudgmentThis type tests the agent's ability to apply the five conversation-type categories at their correct thresholds across a range of ambiguous items. The skill must teach the agent to reason from signal strength and topic substance to the right classification, not to default to the nearest-sounding label or to the manager's stated intent.1
Type D: Multi-Item ReclassificationThis type tests whether the agent can identify and correct multiple misclassified items in a single brief, naming the specific signal that drives each reclassification. The skill must teach the agent to treat reclassification as a systematic signal-reading step applied to every item, not as an edge-case correction applied only when a mismatch is obvious.1
Type E: Reclassification with Recognition HandlingThis type tests whether the agent correctly reclassifies at least one item while also handling the recognition opener appropriately given the available signals. The skill must teach the agent to run both checks independently: reclassification is driven by contradicting evidence, and recognition presence or absence is driven by whether a genuine positive signal exists.1
Type F: Reclassification When Coaching Signal Is AbsentThis type tests whether the agent can reclassify an item the manager has labelled as coaching when the input signals do not support a coaching conversation. The skill must teach the agent that the coaching label requires a behavioural pattern or development signal in the input, and that without one, the item should be reclassified and the reason stated explicitly in the brief.1
Type G: Wrong-Skill RedirectThis type tests whether the agent recognises that the input does not call for a 1-on-1 prep brief and responds with a short, accurate redirect to the correct artefact. The skill must teach the agent to identify the distinguishing characteristics of wrong-skill inputs, such as post-meeting transcripts, skip-level prep requests, or performance-review drafts, and to explain clearly what the right next step is without producing a brief.1

Surface details such as names, team structures, ticket numbers, velocity figures, and topic labels rotate across variants while the underlying capability being tested remains the same.

Datasets

Available Datasets

When you submit, we run your skill through a standardised base agent provided by sundae_bar. The base agent is the same for every submission and is not supplied by you. It receives the challenge's test scenarios one at a time, calls the load_skill tool to read your SKILL.md and the list of your reference files, opens a reference with read_reference when your SKILL.md tells it to, and follows the skill to answer each scenario.

Same base agent, same tests, only the submitted skill changes. Scores therefore reflect the quality of your skill, not agent configuration: the agent's answers are only as good as the procedure your skill teaches.

Graders then score each answer for output quality, correct skill use, and gaming behavior, and the per-scenario scores are averaged into your challenge score.

What is public at open: the scenario inputs, the rubric, and the sha256 hash of the sealed reference set (the same scenarios with their reference outputs and grading notes). Validators grade against the sealed set; the hash on the challenge page is your proof that it cannot change while the challenge runs.

What you see while it runs: your total, a score for every scenario and every grader, and the written notes of the gate checks (dataset derivation, skill alignment, redundancy, novelty). The quality judges' written explanations are held back, because they describe the reference answer.

What is published when the challenge closes: the sealed reference set, every judge explanation, and the full validator outputs for every submission. Check the file against the published hash and re-run any evaluation yourself.

What you submit: a folder or a .zip with SKILL.md at the top level and an optional references/ folder. A single SKILL.md is also accepted. All files are the submission.

Limits: SKILL.md is at most 10,000 bytes and is always loaded. references/ holds up to 30 files, each at most 32,000 characters, 128,000 characters in total. Text formats only: .md, .txt, .json, .yaml, .yml and .csv. Lower-case file names using letters, digits, dots, dashes and underscores. The current values are published by the platform API at GET /api/v2/submissions/limits; the submission form reads them from there.

How references are read: the agent always sees SKILL.md plus a list of your reference file names. It opens a reference only when SKILL.md tells it to, by name. Each read is one harness step within a cap of 10 steps per scenario, and the file's text counts against the scenario's token budget of 17,000 tokens. A reference the agent never opens does nothing for the score.

Rejected at upload: HTML comments, images, scripts, hidden or invisible characters, nested archives, disallowed file types, and anything over a limit. Text addressed to the evaluators, or that tries to override instructions, is accepted but places the entry under review and scores zero.

Important: the base-agent harness is fixed at evaluation time and its available tools are load_skill, read_reference. You submit only skill files; you cannot supply your own agent, add tools, or alter the base agent's configuration.

Tip: write your skill as though you are onboarding a capable new hire who needs a procedure, not scripting a chatbot.

Evaluation Criteria

Each scenario is graded by an LLM judge. Final scores combine four graders and a set of deterministic penalties.

CriteriaWeight

Response quality

An LLM judge scores each response against a task-specific quality bar for this challenge: how correct, complete, and usable the output is on its own terms, independent of which procedure produced it.

30 %

Skill use

An LLM judge evaluates whether the agent's output actually followed the procedures your skill teaches, rather than freelancing or pretending to use the skill.

30 %

Scenario quality

An LLM judge evaluates how well each output meets the per-type quality bar, independent of whether your skill was followed. A reference good output anchors the judge's sense of the quality bar.

30 %

Output originality

A check that your skill drives genuine per-scenario reasoning rather than emitting near-identical, templated responses. Skills that reuse one canned output across scenarios lose credit here.

10 %

Scores are computed at the scenario level and averaged up to capability-type scores, which combine into the final challenge score shown on the leaderboard. Deterministic penalties apply when the agent's output damages quality in ways that are unambiguous, such as inventing signals that were not in the input, producing a full brief for a wrong-skill input, or omitting required structural elements. These penalties are applied before any qualitative assessment, so procedural correctness is a floor, not a tiebreaker.

Submission

How to submit

1
Login or create a sundaebar.ai account

Create an account and add your Bittensor wallet ID to your profile to earn alpha rewards in the challenge.

2
Review the requirements

Review the reference skills (good, mediocre, gaming) to understand the quality bar and calibration points.

3
Build your skill

Draft your SKILL.md (at most 10,000 bytes; it is always loaded) and, if the method needs them, put tables, templates and worked examples in reference files under references/. Tell the agent in SKILL.md which reference to open and when, by file name. Focus on teaching procedure rather than providing templates.

4
Test locally

Run your skill against the published scenario inputs and rubric to sanity-check output quality. The reference outputs are sealed until the challenge closes.

5
Submit via the platform

Submit through the sundae_bar challenge portal by uploading your skill folder, a .zip of it, or a single SKILL.md. The folder has SKILL.md at the top level and an optional references/ folder beside it.

6
Iterate

Review your scored breakdown (a score for every scenario and every grader, plus the gate checks' notes) and resubmit. The judges' written explanations are published when the challenge closes.

Constraints

Submission format: a folder or .zip with SKILL.md (YAML frontmatter with name and description, then a Markdown body) at the top level and an optional references/ folder. A single SKILL.md is also accepted.

Limits: SKILL.md at most 10,000 bytes. Up to 30 reference files, each at most 32,000 characters and 128,000 characters in total; text formats only (.md, .txt, .json, .yaml, .yml and .csv); lower-case file names.

Reference reads: the agent opens a reference only when SKILL.md names it. Each read is one of 10 steps per scenario and its text counts against the scenario's 17,000-token budget.

Base agent: fixed. Provided by sundae_bar.

Runtime: a fixed base-agent harness loads your skill and answers all 8 scenarios per submission with it.

Tests per submission: 8 (the full scenario set, no sampling).

Submission limits: 7 submissions per day. A rolling challenge limit also applies.

Rewards

Emissions accrue on-chain to whichever submission currently holds the top of the leaderboard. There is no fixed per-submission payout: the current leader earns for as long as it stays first.

Time at the top takes all. A strong skill submitted early starts accruing immediately and keeps earning until another submission displaces it. Once you are displaced, accrual stops until you reclaim the top.

Rankings shift as others iterate. Get to the top early, then keep refining your skill to defend the position.

Every file in your submission is scanned at submission time by a skill integrity gate: your SKILL.md and every reference file, whether or not your skill points at it. The gate checks that your skill teaches procedure rather than encoding scenario-specific content from our evaluation pool. Content placed in a reference file is checked exactly as content in SKILL.md.

A skill that fails the integrity gate scores 0 for the submission. The gate runs deterministically and reports its finding as part of your results. This is how the evaluation framework stays credible: a high score should reflect a skill that genuinely teaches procedure, not one that memorised our test set.

Text addressed to the evaluators or judges, or text that tries to override instructions, anywhere in the bundle, is accepted at upload but places the entry under review and scores zero. Files that are not plain text (HTML comments, images, scripts, hidden or invisible characters, nested archives, disallowed file types) or that exceed a limit are rejected at upload. The copy and similarity checks run over the whole bundle, chunk by chunk, so padding a copied skill with extra reference files does not pass them.