VincenzoImp/academic-research-skills / skills/explore-sota
explore-sota
VerifiedUse when building or expanding the SOTA — from an idea, keywords, seed papers, or the existing collection — through MCP search, citation chasing, triage, and digestion of accepted papers.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 3.8 KB |
| references/citation-chasing.md | markdown | 1.7 KB |
references/citation-chasing.md
View on GitHub# Citation Chasing & Anti-Echo-Chamber
## Review scales
| Scale | Use | Included papers | Chasing |
|---|---|---|---|
| quick-scan | orient an idea | ~8–15 | seeds + one high-yield hop |
| focused-sota | defensible related work | ~20–40 | both directions to local saturation |
| full-survey | survey-grade coverage | 50+ | iterative frontier expansion with recorded saturation |
Budgets are planning aids. Stop at saturation or at budget with blind spots
recorded — never pad with weak papers.
## Frontier discipline
1. Expand the current seed set one hop at a time (backward = references,
forward = citers; both via semantic-scholar).
2. Deduplicate before triage: DOI, arXiv id, S2 id, then
title + first author + year.
3. Triage chased candidates with the same inclusion/exclusion criteria as
keyword hits.
4. Promote a candidate into the next seed set only deliberately — when it
is digested and central to the scope. Never recursively chase everything.
5. Saturation = a new hop yields mostly duplicates or out-of-scope records.
6. If the budget ends first, record the unexpanded frontier explicitly
instead of claiming coverage.
## Anti-echo-chamber checks (run before declaring saturation)
- Seeds span at least project-specific, seminal, and recent-frontier types.
- At least one adjacent-terminology or adjacent-community query ran with
terms learned from full readings.
- At least one negative/contrastive query ran (failure, limitation,
negative result, reproduction, replication).
- A sample of rejected near-misses was re-inspected to confirm the criteria
are not too narrow.
- No single author group, venue, benchmark, or citation cluster dominates
the SOTA without a recorded reason.
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: 323k | 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: 323k | 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: 323k | Updated | Claude CodeCodex | Verified | Install |