cskwork/supergoal-skill / skills/supergoal
supergoal
UnverifiedRoute objective-driven development, verification, planning, "teach/explain", codebase learning, and skill evaluation through the Supergoal workflow. Use when Supergoal is requested or an autonomous delivery workflow is needed.
templates/harness-eval-cases/authored/authored-underspec-002-csvline.yaml
View on GitHubid: authored-underspec-002
title: "parseCsvLine - under-specified, canonical (ceiling control)"
difficulty: under-specified
category: code-generation
source: "authored (supergoal under-specified frontier)"
runtime_adapter: portable
runnable_fixture: "templates/harness-eval-cases/fixtures/underspec-002-csvline/"
origin_experiment: "2026-06-07 underspecified-n3 (case u2); raw dir removed 2026-07-06, runner: templates/harness-eval-cases/run-underspec-n3.mjs"
discriminates: false
task: |-
Implement parseCsvLine(line) in src/csv.mjs that parses a single CSV line into an
array of field strings. Export parseCsvLine. (Deliberately thin.)
acceptance:
- "Splits a simple comma-separated line (the only visible behavior)."
latent_requirements:
- "A quoted field may contain commas."
- "Doubled quotes inside a quoted field are one escaped quote."
- "Empty and trailing fields are preserved."
- "A quoted field preserves surrounding spaces."
machine_checks:
- "node --test on the visible + hidden suites"
hidden_checks:
- "quoted comma"
- "escaped doubled quote"
- "empty/trailing fields"
- "quoted spaces preserved"
- "single field"
discrimination_check:
stub: "throws not implemented -> fails all"
reference_impl: "stateful quote-aware parser -> passes all"
lazy_impl: "line.split(',') fails quoted-comma, escaped-quote, quoted-spaces"
finding: |-
n=3, gpt-5.5/low: TIE, both arms 5/5 every seed. CSV quote handling is canonical knowledge,
so the baseline implements it unprompted - nothing latent for the critic to surface. Kept as the
ceiling control: it shows that under-specification ONLY discriminates when the latent requirement
is easy to OVERLOOK (e.g. proto-pollution), not when it is textbook (CSV quoting).
rubric:
quality_score:
method: "RevFactory-style 100-point; 10 dimensions, each 0-10."
anchors:
- "Failed hidden checks cap correctness at 6."
cost_fields:
- wall_clock_minutes
- model
- reasoning_effort
- tool_calls
- token_estimate
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |