Customer Success Account Intelligence Skill
Overview
This skill teaches the agent to read a pool of recent news events alongside an account's contract details and decide — for each event — whether it is a genuine renewal risk, an expansion signal, or noise that should be discarded.
Given a structured account brief (account profile, product scope, champion and buyer roles, contract shape, and a list of pasted news headlines or article summaries), a good result is a clean triage table covering only the events that survived a proximity test, each row explaining exactly which contract element is touched and what action is warranted — plus a single counted summary of everything discarded.
The submission is a single SKILL.md file. Strong skills teach the agent a reasoning procedure it can apply to any account brief it has never seen before: how to run a proximity test, how to classify what survives, how to write account-specific reasoning, and when to refuse or hedge. Weak skills provide fill-in-the-blank templates or lists of example verdicts — these break the moment the account name, industry, or event pool changes.
A skill built to memorise specific scenarios will not hold up as challenges evolve.
You submit a SKILL.md file. A fixed base agent loads it and answers 8 sampled 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 four capability types. Every submission sees 2 randomly sampled variants per type, for 8 total scenarios per evaluation.
| Category | Overview | Test cases |
|---|---|---|
| Type A — Missing-Context Handling | This type tests whether the agent correctly refuses to produce relevance verdicts when required contract context fields are absent, and asks exactly once for the specific missing information. It also tests what happens when the user pushes the agent to proceed anyway: verdicts must be marked provisional and all proximity reasoning must be flagged as unverified. The skill must teach the agent to treat missing context as a blocking condition, not a minor gap to work around. | 2 |
| Type B — Escalation Threshold Discipline | This type tests whether the agent applies the correct fuse-length standard when assigning action tiers. Scenarios present events that are contract-proximate but vary in timing, directness, and urgency relative to the renewal window. The skill must teach the agent to reserve escalate for events with a direct, near-term vector hit inside the renewal window or 90 days, and to assign monitor — not escalate — to events that pass the proximity test but fall outside that threshold. | 2 |
| Type C — Dual-Signal Classification | This type tests whether the agent can correctly handle events that simultaneously carry renewal-risk and expansion-signal properties. The agent must select one act-first relevance class and note the secondary class within the proximity reasoning, rather than splitting the row, omitting one signal, or creating a hybrid label. The skill must teach the agent the procedure for resolving this conflict decisively while preserving full signal visibility. | 2 |
| Type D — Newsworthiness vs. Account Impact | This type tests whether the agent can correctly discard high-profile but contract-irrelevant headlines — such as competitor funding announcements, peer-company regulatory fines, or unrelated product launches — that touch none of the three proximity vectors for the account in question. Scenarios include events that are objectively significant news but should produce a discard verdict and appear only in the counted summary. The skill must teach the agent to anchor every classification decision to the account's specific contract, not to a headline's general market significance. | 2 |
Surface details such as account names, industries, ARR figures, champion roles, renewal dates, and event headlines rotate across variants while the underlying capability being tested remains the same.
Datasets
Available Datasets
When you submit, we run your SKILL.md 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, and follows it 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.
Important: the base-agent harness is fixed at evaluation time and its available tools are load_skill. You submit only a SKILL.md; 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.
| Criteria | Weight | |
|---|---|---|
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 % |
Scenarios are scored individually, and deterministic penalties apply when agent outputs damage output quality — for example by carrying noise events into the triage table, issuing verdicts without proximity reasoning, misapplying action tiers, or producing account-agnostic reasoning that could belong to any brief. Individual scenario scores roll up to capability-type averages, which combine into the final challenge score shown on the leaderboard.
Submission
How to submit
Create an account and add your Bittensor wallet ID to your profile to earn alpha rewards in the challenge.
Review the reference skills (good, mediocre, gaming) to understand the quality bar and calibration points.
Draft your SKILL.md. Focus on teaching procedure rather than providing templates.
Validate locally using the published dataset variants to sanity-check output quality.
Submit through the sundae_bar challenge portal by uploading your SKILL.md file.
Review your scored breakdown and resubmit.
Constraints
Submission format: a single SKILL.md file with YAML frontmatter (name, description) and Markdown body.
Base agent: fixed. Provided by sundae_bar.
Runtime: a fixed base-agent harness loads your SKILL.md and answers 8 sampled scenarios per submission with it.
Tests per submission: 8 (2 random variants per capability type, seeded by task ID for reproducibility).
Submission limits: 3 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.
Your SKILL.md is scanned at submission time by a skill integrity gate. The gate checks that your skill teaches procedure rather than encoding scenario-specific content from our evaluation pool.
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.