Skip to content

hashgraph-online/awesome-codex-plugins / plugins/beefiker/superloopy/skills/superloopy-research

superloopy-research

Verified

Use only after explicit Codex `$superloopy:superloopy-research` or Claude Code `/superloopy:superloopy-research` invocation, a research task started with a leading `loopy` or `루피` (such as `loopy research`), or an active Superloopy loop explicitly routing a research deliverable here. Evidence-backed Superloopy research orchestration with automatic advisory usage targets, parallel read-only lanes, per-retrieval verdicts, graded and dated sources, empirical verification, a priced claim ledger, cited synthesis, and optional reports. Do not activate from research, investigate, look-up, summarize, deep-dive, report, or similar vocabulary alone, in any language; ordinary questions, debugging, and implementation context-gathering stay with their primary workflows.

FileKindSize
SKILL.mdskill md14.8 KB
agents/openai.yamlconfig410 B
references/claims.mdmarkdown6.0 KB
references/lanes.mdmarkdown8.9 KB
references/retrieval.mdmarkdown7.6 KB
references/scope.mdmarkdown3.7 KB
references/synthesis.mdmarkdown4.2 KB
scripts/validate-research-evidence.mjsscript23.4 KB

references/synthesis.md

View on GitHub
# Synthesis

Load this reference for Phase 4 and Phase 5. The completion checklist and the mechanical gate stay in `SKILL.md`.

## Phase 4 - Synthesize

After convergence and all verifications, read `INDEX.md` first, then open only the wave, verify, and ledger files the themes you are actually writing cite. The index exists so that synthesis does not require pulling the whole journal back into context; go to a wave file when a specific claim needs its detail, not by default. Then write `SYNTHESIS.md`:

```text
# Superloopy Research Synthesis: <query>
Workers: <total> · Waves: <count> · Sources: <count> · Verifications: <count>
Research usage: workers <observed>/<target> · queries <observed-or-unknown>/<target> · waves <observed>/<target> · provenance <handoff|journal|self-reported>
Retrieval integrity: ok <n> · partial <n> · blocked <n> · empty <n> · quota <spent-or-unknown> · re-dispatched lanes <n>

## Executive answer        — 2-3 paragraphs answering the core question, valid as of <date>
## Findings by theme       — per theme: consensus, evidence links, key quote (<20 words, attributed), verified yes/no
## Codebase findings       — absolute or repo-relative paths with line references
## Sources (ranked)        — URL, what it contains, grade, retrieval verdict, observed date, as-of date
## Verified claims         — code: claim | verdict | verify-<slug>.md · non-code: only rows cleared into verified-claims
## Contradictions          — source A vs source B, which surfaces they sit on, resolution with evidence
## Gaps                    — what saturation could not answer · unresolved/refuted claim-ledger rows · blocked sources with no substitute · deferred claims and why
## Expansion trace         — per wave: workers → markers; retrieval verdict mix; convergence reason
```

Deliver the synthesis with inline `[Source N]` citations on every substantive claim. In `## Sources`, define each numbered source on its own bullet as `- Source N: <locator> ...`; prose elsewhere never defines a source. In `## Verified claims`, put every claim on a structured row `- <claim-id> | <verdict> | <artifact-or-ledger>` so the validator can reject uncleared ledger ids and require every non-ledger id to name a present `verify-<slug>.md` artifact. Every high-risk non-code claim you assert must be a verified-claims row from Phase 3b — assert nothing left in the unresolved/refuted annex. Keep direct quotes short and attributed; do not copy long passages. When no report was requested, this is the deliverable.

## Phase 5 - Report (only when requested)

Format by the user's words: "report"/"document" → markdown (default) · "pdf" → HTML first, then a renderer · "slides"/"presentation"/"deck" → a slide builder · "html"/"webpage" → standalone HTML.

Asset workers (parallel): charts for quantitative findings, full-page screenshots of the top 5-10 sources, and generated diagrams when architecture or flows need them — saved by you under `<evidence-root>/assets/`.

Assembly: before writing, load every available design and visualization skill and apply it — the report is a designed artifact, not a text dump. Structure: executive summary → key findings by theme → detailed analysis (quotes under 20 words with attribution, charts, SHA-pinned permalinks, verification results) → comparative analysis when options compete → numbered sources with access dates → methodology appendix (workers, waves, searches, verifications). Every claim cites `[Source N]`. The orchestrator owns this write: assemble it yourself, or have a writing lane draft content returned as message text and write it under the evidence root — a designated writing worker that produces the file ends its reply with `SUPERLOOPY_EVIDENCE: <path-under-active-evidence-root>`.

Close the research with `superloopy loop evidence --status pass --artifact <report-or-synthesis>` pointing at the deliverable. Note: `superloopy loop report` is a *separate* command that generates a complementary evidence-trace summary (evidence root, ledger, progress) to its own path — it is not a publisher for this designed report, so never point it at your report file or it will overwrite your content. Run it, if at all, against a distinct path such as `<evidence-root>/evidence-report.md`.
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: 471kUpdated Claude CodeCodexVerifiedInstall
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: 322kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 322kUpdated Claude CodeCodexVerifiedInstall
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 availableStars: 322kUpdated Claude CodeCodexVerifiedInstall