{"id":"4666c3fb-e4cf-4c3d-9c82-3ed0173a8ee4","shortId":"7gq7pz","kind":"skill","title":"reaper","tagline":"Run the full Reaper research pipeline: analyze a paper (optional), review literature, formalize the problem, investigate hypotheses with critique loops, and synthesize a report. Use when given a research goal, optionally with a paper.","description":"# Reaper: AI-Native Research Pipeline\n\nYou are the Reaper orchestrator. You take a research goal — optionally with a research paper — and autonomously conduct rigorous, multi-step academic research.\n\n## Usage\n\nInvoke the `/reaper` skill with a research goal (quoted string), optionally followed by a paper path and/or `--codex`. Examples below use a generic name-based form; the slash-command form `/reaper \"...\"` works on hosts with slash-command routing (e.g. Claude Code).\n\n```\n# Without a paper — pure goal-driven research\nreaper \"explore the feasibility of post-quantum threshold signatures\"\n\n# With a paper\nreaper \"determine if the security proof in Section 4 holds under asynchrony\" path/to/paper.pdf\n\n# With external-model consultation for automated AI feedback between investigation cycles\nreaper \"determine if the security proof in Section 4 holds under asynchrony\" path/to/paper.pdf --codex\n```\n\n**Argument parsing:** The research goal (quoted string) is required. If a path to an existing file (PDF or text) is also provided, treat it as the paper. Pass `--codex` to enable external-model consultation across the entire pipeline — every skill gains an optional step where it consults an external model for a second opinion at a natural checkpoint. See `references/codex-consultation.md` for the full protocol. Requires a host with MCP support and a registered Codex MCP server (e.g. [codex-mcp-server](https://github.com/tuannvm/codex-mcp-server)); silently no-op on hosts without MCP.\n\nWhen `--codex` is set, propagate this context to all skill invocations. Each skill will consult Codex only at its designated checkpoint (defined in `references/codex-consultation.md`), using compressed context and a shared session ID for continuity across the pipeline.\n\n## Peer Skills\n\nThis orchestrator chains 8 sub-skills that must be installed alongside it: `/clarify-goal`, `/analyze-paper`, `/review-literature`, `/formalize-problem`, `/brainstorm`, `/investigate`, `/critique`, `/synthesize`. One more (`/search-paper`) is called transitively by `/review-literature` and `/investigate`. The `/<skill>` form is the canonical display convention used in these docs; substitute the host-native invocation form (slash command, auto-discovery, manual `SKILL.md` pointer) when actually running them.\n\nIf any of these are missing from your agent's skills folder, ask the user to reinstall the full Reaper package (`npx skills add SebastianElvis/reaper`).\n\n## Invocation Convention\n\nReferences below to running a sub-skill use the host-agnostic phrase \"invoke the `<name>` skill\" — invoke each sub-skill by its `name` using your host's native skill-loading mechanism. The loaded `SKILL.md` provides the full instructions for that stage. Concrete invocation form varies by host:\n\n- **Slash-command hosts** (e.g. Claude Code): `/<sub>` (e.g. `/clarify-goal`)\n- **Auto-discovery hosts** (e.g. Cursor, Codex CLI, Cline, Continue, Gemini CLI, Copilot, Windsurf, OpenCode): the agent loads peer `SKILL.md` files from the skills folder and routes by `name` + `description` match.\n- **Manual invocation hosts**: explicitly point the agent at the installed skill's `SKILL.md` (typical paths: `~/.claude/skills/<name>/SKILL.md`, `~/.cursor/skills/<name>/SKILL.md`, `~/.agents/skills/<name>/SKILL.md`, `~/.continue/skills/<name>/SKILL.md`, `~/.windsurf/skills/<name>/SKILL.md`, or `<repo-root>/skills/<name>/SKILL.md` during development — substitute `<name>` with the sub-skill directory name like `clarify-goal`).\n\n## Design Principles\n\n1. **Separation of concerns**: Each skill has one job. Skills communicate through workspace files, not shared memory.\n2. **Fixed evaluation signal**: Every investigation cycle produces a binary keep/discard verdict.\n3. **Structured results log**: `notes/results.md` is the single source of truth for what has been tried and what happened.\n4. **Keep-or-discard loop**: Only \"keep\" findings flow into `current-understanding.md`. Discards stay in the audit trail.\n5. **Never stop**: The pipeline runs to completion. Uncertainty about user intent is never a reason to pause.\n6. **Clarity and simplicity**: Prefer simpler proofs, fewer assumptions, shorter arguments.\n\n## Workflow\n\n### Step 1: Initialize Workspace\n\nCreate the workspace directory structure:\n\n```\nreaper-workspace/\n├── notes/                          # Evolving — edited inline to reflect latest state\n│   ├── current-understanding.md    # Initialize empty — \"branch tip\"\n│   └── results.md                  # Initialize with table header — one row per hypothesis, updated inline on revisit\n├── papers/                         # Evolving — per-paper notes can be updated\n├── investigations/                 # Evolving — reuse directory on revisit, edit inline\n├── feedbacks/                      # Append-only — one file per event, never modified\n├── logs/                           # Append-only — one file per cycle, never modified\n```\n\n**File mutation rules:**\n\n| Category | Files | Policy |\n|----------|-------|--------|\n| Evolving (inline edit) | `notes/current-understanding.md`, `notes/results.md`, `notes/ideas.md`, `notes/problem-statement.md`, `notes/literature.md`, `papers/*-notes.md`, `investigations/*/analysis.md`, `investigations/*/proof.md` | Edit in place. Single writer per file per batch. `notes/literature.md` is created by `/review-literature` and extended inline by `/investigate` during mid-cycle literature search. |\n| Append-only | `logs/cycle-*.md`, `feedbacks/round-*.md`, `feedbacks/codex-consultation-*.md` | Create once, never modify. |\n| Write-once | `notes/paper-summary.md`, `notes/clarified-goal.md`, `report.md` | Created by one skill. May be regenerated on re-run but not incrementally edited by other skills. |\n\n**File naming conventions:** Investigation dirs: `NNN-<slug>/` (zero-padded). Cycle logs: `cycle-NNN-<slug>.md`. Feedback: `round-N.md`, `codex-consultation-N.md`. Paper notes: `<id>-notes.md`.\n\n**Lazy-load protocol:** Early-pipeline skills (`/formalize-problem`, `/analyze-paper`) read source files eagerly. Loop skills (`/brainstorm`, `/investigate`, `/critique`, `/synthesize`) use `current-understanding.md` as primary source and lazy-load `paper-summary.md` and `literature.md` only when stuck or when a specific hypothesis requires it.\n\nInitialize `reaper-workspace/notes/results.md` with:\n```markdown\n# Investigation Results\n\n| Cycle | Hypothesis | Action | Outcome | Confidence | Status | Description |\n|-------|-----------|--------|---------|------------|--------|-------------|\n```\n\nInitialize `reaper-workspace/notes/current-understanding.md` with:\n```markdown\n# Current Understanding\n\n*No investigation cycles completed yet.*\n```\n\n### Step 2: Clarify the Research Goal\n\n**If a paper was provided:**\nInvoke the **`/clarify-goal`** skill with arguments `\"<research-goal>\" <paper-path>` — it does a quick scan of the paper, asks the user 3-5 targeted clarifying questions about scope, assumptions, and success criteria, then writes `notes/clarified-goal.md`.\n\n**If no paper was provided:**\nInvoke the **`/clarify-goal`** skill with argument `\"<research-goal>\"` — it asks the user 3-5 targeted clarifying questions based on the goal alone (no paper scan), then writes `notes/clarified-goal.md`.\n\nIf the goal is already precise and unambiguous, this step writes the file without asking questions.\n\nAll downstream skills should read `clarified-goal.md` for the refined goal and context.\n\n### Step 3: Establish Baseline (parallel)\n\n**If a paper was provided**, run these two skills as **parallel subagents** using your host's parallel-spawn primitive (e.g. Claude Code's `Agent` tool, or the equivalent on your host; if the host has no parallel primitive, run them sequentially) — they write to non-overlapping files:\n\n1. Invoke **`/analyze-paper`** with `<paper-path>` — produces `notes/paper-summary.md`\n2. Invoke **`/review-literature`** with `\"<research-goal>\"` — produces `notes/literature.md`\n\nUse the refined goal from `clarified-goal.md` for the literature review argument. Both must complete before proceeding.\n\n**If no paper was provided**, run only:\n\n1. Invoke **`/review-literature`** with `\"<research-goal>\"` — produces `notes/literature.md`\n\nSkip `/analyze-paper` entirely — there is no paper to analyze. The pipeline proceeds without `notes/paper-summary.md`.\n\n### Step 4: Formalize the Problem\n\nInvoke **`/formalize-problem`** with `\"<research-goal>\"` — it reads the baseline outputs (`clarified-goal.md`, `literature.md`, and `paper-summary.md` if available) and produces `notes/problem-statement.md` (trust assumptions, security properties, performance goals) and `notes/ideas.md` (prioritized ideas).\n\n### Step 5: Brainstorm → Investigate → Critique Loop\n\nRun the core research engine: a recurring loop of brainstorm (generate ideas), investigate (execute on ideas), and critique (external perspective).\n\n#### Adaptive Loop Structure\n\nThe loop adapts to problem complexity. Assess complexity after formalization, before the first brainstorm:\n\n- **Simple** (1-3 hypotheses, narrow scope, single claim to check): 2 brainstorm-investigate rounds, 3 cycles each, 1 critique. Total: ~6 investigation cycles.\n- **Moderate** (3-5 hypotheses, multiple steps or properties): 2 brainstorm-investigate rounds, 5 cycles each, 1-2 critiques. Total: ~10 investigation cycles.\n- **Complex** (5+ hypotheses, multiple interacting properties, broad scope): 3+ brainstorm-investigate rounds, 5 cycles each, 2+ critiques. Total: ~15+ investigation cycles.\n\n#### Default Patterns\n\n**Default (no `--codex`):**\n```\nbrainstorm  →  investigate N  →  critique --self  →  brainstorm  →  investigate N\n```\n\n**With `--codex`:**\n```\nbrainstorm  →  investigate N  →  critique --codex  →  brainstorm  →  investigate N  →  critique --codex\n```\n\nWhere N is determined by the complexity assessment above.\n\n#### Adaptation Signals During the Loop\n\n- If >50% of cycles in a batch are \"discard,\" the hypotheses may be poorly formulated — run brainstorm before the next batch rather than continuing with existing hypotheses.\n- If investigate returns early (all hypotheses resolved), run brainstorm to check for overlooked directions before moving to synthesis.\n- If a critique round generates 3+ new actionable hypotheses, extend the loop with another investigate batch.\n- If 2 consecutive batches produce only \"keep\" results with high confidence and no new hypotheses, convergence is likely — proceed to synthesis.\n\n#### Re-Formalization Handling\n\nIf `/investigate` returns with a re-formalization signal (any cycle logged with outcome `reformulate`):\n\n1. **Preserve state**: Do NOT clear `notes/results.md`, `notes/current-understanding.md`, or existing investigation directories.\n2. **Archive old formulation**: Rename `notes/problem-statement.md` to `notes/problem-statement-v<N>.md` before re-running.\n3. **Re-run `/formalize-problem`**: It reads the reformulation signal from the triggering cycle's `analysis.md`.\n4. **Reset ideas selectively**: In `ideas.md`, mark hypotheses that depended on the old formulation as `[superseded by v<N>]`. Add new hypotheses below.\n5. **Restart loop**: Fresh cycle budget for the new formulation.\n\n#### Loop Mechanics\n\nThe `/brainstorm` step reads the current state and updates `ideas.md` (adding new ideas, editing existing ones inline) (tagged `[Brainstorm-N]`). The `/critique` step may also add hypotheses (tagged `[Codex-N]` or `[Self-N]`). The next `/investigate` batch picks up all unresolved ideas automatically.\n\nThis loop runs autonomously — do not interrupt or ask if it should continue.\n\n### Step 6: Synthesize\n\nInvoke the **`/synthesize`** skill — it reads all workspace files and produces `report.md`.\n\n### Step 7: Present Results\n\nAfter synthesis completes:\n1. Read `reaper-workspace/report.md`\n2. Present the key findings to the user\n3. Point them to the full workspace for the audit trail\n4. Proceed to Step 8\n\n### Step 8: Offer Iteration\n\nAfter presenting results, let the user know they can iterate:\n\n> If you'd like to refine, deepen, or challenge any aspect of this research, invoke the `/critique` skill with your feedback as a quoted string. (Slash-command hosts: `/critique \"your feedback here\"`.)\n\nDo **not** block waiting for a response — the pipeline is complete. The user can invoke `/critique` with quoted feedback at any time to start a feedback round. The critique skill classifies the feedback, may run targeted investigation cycles, and then you should re-invoke `/synthesize` to produce an updated report.\n\n## Skill Dependency Graph\n\n```\n/clarify-goal ──► /analyze-paper ──┐\n             (if paper provided)   ├──► /formalize-problem ──► /brainstorm ◄──► /investigate ◄──► /critique\n                /review-literature ┘                                                              │\n                  (calls /analyze-paper                                     /synthesize ◄──────────┘\n                   for each paper)\n```\n\n| Skill | Requires | Produces |\n|-------|----------|----------|\n| `/clarify-goal` | (paper path, optional) | `notes/clarified-goal.md` |\n| `/analyze-paper` | (paper path) — **skipped at top level if no paper**; also called by `/review-literature` for each downloaded paper | `notes/paper-summary.md` or `papers/<id>-notes.md` (when `--output` is specified) |\n| `/review-literature` | (research goal); calls `/analyze-paper --goal` per downloaded paper | `notes/literature.md`, `papers/*` |\n| `/formalize-problem` | `clarified-goal.md`, `literature.md`, `paper-summary.md` (optional) | `problem-statement.md`, `ideas.md` |\n| `/brainstorm` | `problem-statement.md`, `ideas.md`, `current-understanding.md`, `results.md` | Updates `ideas.md` |\n| `/investigate` | `problem-statement.md`, `ideas.md`, `current-understanding.md`, `results.md` | Updates `results.md`, `current-understanding.md`, `ideas.md`; creates `investigations/*`, `logs/*` |\n| `/critique` | `current-understanding.md`, `results.md`, `problem-statement.md`, `ideas.md` | `feedbacks/*`; may update `ideas.md` |\n| `/synthesize` | `current-understanding.md`, `results.md`, `problem-statement.md`, `ideas.md` | `report.md` |\n| `/search-paper` | (query) | stdout — search results, citations, or resolved venue |\n\n## Important Notes\n\n- Sub-skills are invoked using the host agent's native skill mechanism — by `name` plus arguments. The exact API differs per host (e.g. Claude Code's `Skill` tool with `skill: \"analyze-paper\", args: \"paper.pdf\"`; Cursor/Codex/Cline auto-route based on the loaded `SKILL.md`). Refer to the host's skill documentation for the exact form.\n- Skills communicate exclusively through workspace files — no in-memory state passing\n- If a skill fails, read its output file to diagnose, then retry\n- The workspace is the source of truth — if context is compressed, re-read workspace files\n- Never ask the user \"should I continue?\" during the pipeline — run to completion\n\n## Context Window Recovery\n\nIf the context window is compressed or the orchestrator loses track of its position, reconstruct the pipeline state from workspace files alone:\n\n1. **Check which files exist** in `reaper-workspace/notes/`\n2. **Read `notes/results.md`** to determine investigation progress (count cycles, check for unresolved hypotheses)\n3. **Read `ideas.md`** to find unresolved hypotheses (cross-reference with `notes/results.md`)\n4. **Check for `report.md`** to determine if synthesis is complete\n5. **Check `feedbacks/`** for iteration rounds\n\n**Decision table:**\n\n| `paper-summary.md` | `problem-statement.md` | `notes/results.md` rows | `report.md` | Action |\n|---|---|---|---|---|\n| missing | - | - | - | Run Step 3 (baseline). If no paper was provided, this is normal — proceed with literature review only. |\n| exists or N/A | missing | - | - | Run Step 4 (formalize) |\n| exists or N/A | exists | 0 | - | Run Step 5 (brainstorm + investigate). Check `ideas.md` for unresolved hypotheses. |\n| exists or N/A | exists | >0, unresolved H | - | Continue Step 5 (investigate or brainstorm). Check `ideas.md` for unresolved hypotheses. |\n| exists or N/A | exists | >0, all H resolved | missing | Run Step 6 (synthesize) |\n| exists or N/A | exists | >0 | exists | Run Step 7-8 (present + offer iteration) |","tags":["reaper","sebastianelvis","agent-skills","ai-research","arxiv","claude-code","cline","codex-cli","cryptography","cursor","dblp","distributed-systems"],"capabilities":["skill","source-sebastianelvis","skill-reaper","topic-agent-skills","topic-ai-research","topic-arxiv","topic-claude-code","topic-cline","topic-codex-cli","topic-cryptography","topic-cursor","topic-dblp","topic-distributed-systems","topic-gemini-cli","topic-iacr"],"categories":["reaper"],"synonyms":[],"warnings":[],"endpointUrl":"https://skills.sh/SebastianElvis/reaper/reaper","protocol":"skill","transport":"skills-sh","auth":{"type":"none","details":{"cli":"npx skills add SebastianElvis/reaper","source_repo":"https://github.com/SebastianElvis/reaper","install_from":"skills.sh"}},"qualityScore":"0.454","qualityRationale":"deterministic score 0.45 from registry signals: · indexed on github topic:agent-skills · 8 github stars · SKILL.md body (16,187 chars)","verified":false,"liveness":"unknown","lastLivenessCheck":null,"agentReviews":{"count":0,"score_avg":null,"cost_usd_avg":null,"success_rate":null,"latency_p50_ms":null,"narrative_summary":null,"summary_updated_at":null},"enrichmentModel":"deterministic:skill-github:v1","enrichmentVersion":1,"enrichedAt":"2026-05-18T19:08:57.201Z","embedding":null,"createdAt":"2026-05-18T13:14:41.731Z","updatedAt":"2026-05-18T19:08:57.201Z","lastSeenAt":"2026-05-18T19:08:57.201Z","tsv":"'-2':1197 '-3':1158 '-5':900,929,1182 '-8':2039 '/.agents/skills':500 '/.claude/skills':496 '/.continue/skills':502 '/.cursor/skills':498 '/.windsurf/skills':504 '/analysis.md':713 '/analyze-paper':317,808,1028,1068,1653,1663,1676,1706 '/brainstorm':320,815,1441,1658,1720 '/clarify-goal':316,449,884,920,1652,1671 '/critique':322,817,1462,1581,1594,1613,1660,1739 '/formalize-problem':319,807,1087,1394,1657,1713 '/investigate':321,333,734,816,1351,1478,1659,1727 '/notes':1908 '/notes/current-understanding.md':861 '/notes/results.md':845 '/proof.md':715 '/reaper':69,99 '/report.md':1526 '/review-literature':318,331,729,1034,1063,1661,1689,1702 '/search-paper':326,1754 '/skill.md':497,499,501,503,505,508 '/skills':507 '/synthesize':323,818,1504,1643,1664,1748 '/tuannvm/codex-mcp-server));':255 '0':1988,2003,2021,2034 '1':525,622,1026,1061,1157,1174,1196,1365,1521,1899 '10':1200 '15':1222 '2':542,872,1032,1166,1188,1219,1326,1377,1527,1909 '3':554,899,928,973,1171,1181,1211,1314,1390,1535,1922,1961 '4':140,165,573,1082,1406,1546,1934,1982 '5':591,1114,1193,1204,1216,1428,1944,1991,2008 '50':1265 '6':609,1177,1500,2028 '7':1515,2038 '8':306,1550,1552 'academ':64 'across':206,298 'action':852,1316,1957 'actual':361 'ad':1450 'adapt':1139,1144,1259 'add':387,1424,1466 'agent':372,466,487,1001,1773 'agnost':403 'ai':38,152 'ai-nat':37 'alon':937,1898 'alongsid':314 'alreadi':948 'also':191,1465,1686 'analysis.md':1405 'analyz':8,1075,1797 'analyze-pap':1796 'and/or':83 'anoth':1322 'api':1784 'append':678,688,742 'append-on':677,687,741 'archiv':1378 'arg':1799 'argument':171,619,887,923,1048,1781 'ask':376,896,925,958,1494,1862 'aspect':1575 'assess':1148,1257 'assumpt':617,906,1104 'asynchroni':143,168 'audit':589,1544 'auto':355,451,1803 'auto-discoveri':354,450 'auto-rout':1802 'autom':151 'automat':1485 'autonom':58,1489 'avail':1099 'base':92,933,1805 'baselin':975,1092,1962 'batch':724,1270,1284,1324,1328,1479 'binari':551 'block':1600 'brainstorm':1115,1128,1155,1168,1190,1213,1230,1235,1240,1245,1280,1299,1459,1992,2011 'brainstorm-investig':1167,1189,1212 'brainstorm-n':1458 'branch':644 'broad':1209 'budget':1433 'call':328,1662,1687,1705 'canon':338 'categori':699 'chain':305 'challeng':1573 'check':1165,1301,1900,1918,1935,1945,1994,2012 'checkpoint':229,284 'citat':1759 'claim':1163 'clarifi':521,873,902,931 'clarified-goal.md':965,1043,1094,1714 'clarify-go':520 'clariti':610 'classifi':1628 'claud':109,446,998,1789 'clear':1370 'cli':457,461 'cline':458 'code':110,447,999,1790 'codex':84,170,199,245,250,265,279,456,1229,1239,1244,1249,1470 'codex-consultation-n.md':795 'codex-mcp-serv':249 'codex-n':1469 'command':97,106,353,443,1592 'communic':535,1822 'complet':598,869,1051,1520,1608,1873,1943 'complex':1147,1149,1203,1256 'compress':289,1855,1882 'concern':528 'concret':435 'conduct':59 'confid':854,1335 'consecut':1327 'consult':149,205,218,278 'context':270,290,971,1853,1874,1879 'continu':297,459,1287,1498,1867,2006 'convent':340,390,780 'converg':1340 'copilot':462 'core':1121 'count':1916 'creat':625,727,750,760,1736 'criteria':909 'critiqu':20,1117,1136,1175,1198,1220,1233,1243,1248,1311,1626 'cross':1930 'cross-refer':1929 'current':864,1445 'current-understanding.md':584,641,820,1723,1730,1734,1740,1749 'cursor':455 'cursor/codex/cline':1801 'cycl':156,548,693,738,787,790,850,868,1172,1179,1194,1202,1217,1224,1267,1360,1403,1432,1635,1917 'cycle-nnn':789 'd':1567 'decis':1950 'deepen':1571 'default':1225,1227 'defin':285 'depend':1415,1650 'descript':479,856 'design':283,523 'determin':133,158,1253,1913,1939 'develop':510 'diagnos':1842 'differ':1785 'dir':782 'direct':1304 'directori':517,628,671,1376 'discard':577,585,1272 'discoveri':356,452 'display':339 'doc':344 'document':1816 'download':1692,1709 'downstream':961 'driven':117 'e.g':108,248,445,448,454,997,1788 'eager':812 'earli':804,1294 'early-pipelin':803 'edit':635,674,704,716,774,1453 'empti':643 'enabl':201 'engin':1123 'entir':208,1069 'equival':1005 'establish':974 'evalu':544 'event':683 'everi':210,546 'evolv':634,660,669,702 'exact':1783,1819 'exampl':85 'exclus':1823 'execut':1132 'exist':185,1289,1374,1454,1903,1976,1984,1987,1999,2002,2017,2020,2030,2033,2035 'explicit':484 'explor':120 'extend':731,1318 'extern':147,203,220,1137 'external-model':146,202 'fail':1836 'feasibl':122 'feedback':153,676,793,1585,1596,1616,1623,1630,1744,1946 'feedbacks/codex-consultation-':748 'feedbacks/round-':746 'fewer':616 'file':186,470,538,681,691,696,700,722,778,811,956,1025,1510,1826,1840,1860,1897,1902 'find':581,1531,1926 'first':1154 'fix':543 'flow':582 'folder':375,474 'follow':78 'form':93,98,335,351,437,1820 'formal':14,1083,1151,1348,1357,1983 'formul':1278,1380,1419,1437 'fresh':1431 'full':4,234,382,430,1540 'gain':212 'gemini':460 'generat':1129,1313 'generic':89 'github.com':254 'github.com/tuannvm/codex-mcp-server));':253 'given':28 'goal':31,51,74,116,175,522,876,936,946,969,1041,1108,1704,1707 'goal-driven':115 'graph':1651 'h':2005,2023 'handl':1349 'happen':572 'header':650 'high':1334 'hold':141,166 'host':102,238,261,348,402,418,440,444,453,483,991,1008,1011,1593,1772,1787,1813 'host-agnost':401 'host-nat':347 'hypothes':18,1159,1183,1205,1274,1290,1296,1317,1339,1413,1426,1467,1921,1928,1998,2016 'hypothesi':654,838,851 'id':295 'idea':1112,1130,1134,1408,1452,1484 'ideas.md':1411,1449,1719,1722,1726,1729,1735,1743,1747,1752,1924,1995,2013 'import':1763 'in-memori':1828 'increment':773 'initi':623,642,647,841,857 'inlin':636,656,675,703,732,1456 'instal':313,490 'instruct':431 'intent':602 'interact':1207 'interrupt':1492 'investig':17,155,547,668,712,714,781,848,867,1116,1131,1169,1178,1191,1201,1214,1223,1231,1236,1241,1246,1292,1323,1375,1634,1737,1914,1993,2009 'invoc':274,350,389,436,482 'invok':67,405,408,882,918,1027,1033,1062,1086,1502,1579,1612,1642,1769 'iter':1554,1564,1948,2042 'job':533 'keep':575,580,1331 'keep-or-discard':574 'keep/discard':552 'key':1530 'know':1561 'latest':639 'lazi':800,826 'lazy-load':799,825 'let':1558 'level':1682 'like':519,1342,1568 'literatur':13,739,1046,1973 'literature.md':830,1095,1715 'load':423,426,467,801,827,1808 'log':557,686,788,1361,1738 'logs/cycle-':744 'loop':21,578,813,1118,1126,1140,1143,1263,1320,1430,1438,1487 'lose':1886 'manual':357,481 'mark':1412 'markdown':847,863 'match':480 'may':764,1275,1464,1631,1745 'mcp':240,246,251,263 'md':745,747,749,792,1385 'mechan':424,1439,1777 'memori':541,1830 'mid':737 'mid-cycl':736 'miss':369,1958,1979,2025 'model':148,204,221 'moder':1180 'modifi':685,695,753 'move':1306 'multi':62 'multi-step':61 'multipl':1184,1206 'must':311,1050 'mutat':697 'n':1232,1237,1242,1247,1251,1460,1471,1475 'n/a':1978,1986,2001,2019,2032 'name':91,415,478,518,779,1779 'name-bas':90 'narrow':1160 'nativ':39,349,420,1775 'natur':228 'never':592,604,684,694,752,1861 'new':1315,1338,1425,1436,1451 'next':1283,1477 'nnn':783,791 'no-op':257 'non':1023 'non-overlap':1022 'normal':1970 'note':633,664,797,1764 'notes.md':711,798,1697 'notes/clarified-goal.md':758,912,943,1675 'notes/current-understanding.md':705,1372 'notes/ideas.md':707,1110 'notes/literature.md':709,725,1037,1066,1711 'notes/paper-summary.md':757,1031,1080,1694 'notes/problem-statement-v':1384 'notes/problem-statement.md':708,1102,1382 'notes/results.md':558,706,1371,1911,1933,1954 'npx':385 'offer':1553,2041 'old':1379,1418 'one':324,532,651,680,690,762,1455 'op':259 'opencod':464 'opinion':225 'option':11,32,52,77,214,1674,1717 'orchestr':46,304,1885 'outcom':853,1363 'output':1093,1699,1839 'overlap':1024 'overlook':1303 'packag':384 'pad':786 'paper':10,35,56,81,113,131,197,659,663,710,796,879,895,915,939,979,1056,1073,1655,1667,1672,1677,1685,1693,1696,1710,1712,1798,1965 'paper-summary.md':828,1097,1716,1952 'paper.pdf':1800 'parallel':976,987,994,1014 'parallel-spawn':993 'pars':172 'pass':198,1832 'path':82,182,495,1673,1678 'path/to/paper.pdf':144,169 'pattern':1226 'paus':608 'pdf':187 'peer':301,468 'per':653,662,682,692,721,723,1708,1786 'per-pap':661 'perform':1107 'perspect':1138 'phrase':404 'pick':1480 'pipelin':7,41,209,300,595,805,1077,1606,1870,1893 'place':718 'plus':1780 'point':485,1536 'pointer':359 'polici':701 'poor':1277 'posit':1890 'post':125 'post-quantum':124 'precis':949 'prefer':613 'present':1516,1528,1556,2040 'preserv':1366 'primari':822 'primit':996,1015 'principl':524 'priorit':1111 'problem':16,1085,1146 'problem-statement.md':1718,1721,1728,1742,1751,1953 'proceed':1053,1078,1343,1547,1971 'produc':549,1030,1036,1065,1101,1329,1512,1645,1670 'progress':1915 'proof':137,162,615 'propag':268 'properti':1106,1187,1208 'protocol':235,802 'provid':192,428,881,917,981,1058,1656,1967 'pure':114 'quantum':126 'queri':1755 'question':903,932,959 'quick':891 'quot':75,176,1588,1615 'rather':1285 're':769,1347,1356,1388,1392,1641,1857 're-form':1346,1355 're-invok':1640 're-read':1856 're-run':768,1387,1391 'read':809,964,1090,1396,1443,1507,1522,1837,1858,1910,1923 'reaper':1,5,36,45,119,132,157,383,631,843,859,1524,1906 'reaper-workspac':630,842,858,1523,1905 'reason':606 'reconstruct':1891 'recoveri':1876 'recur':1125 'refer':391,1810,1931 'references/codex-consultation.md':231,287 'refin':968,1040,1570 'reflect':638 'reformul':1364,1398 'regener':766 'regist':244 'reinstal':380 'renam':1381 'report':25,1648 'report.md':759,1513,1753,1937,1956 'requir':179,236,839,1669 'research':6,30,40,50,55,65,73,118,174,875,1122,1578,1703 'reset':1407 'resolv':1297,1761,2024 'respons':1604 'restart':1429 'result':556,849,1332,1517,1557,1758 'results.md':646,1724,1731,1733,1741,1750 'retri':1844 'return':1293,1352 'reus':670 'review':12,1047,1974 'revisit':658,673 'rigor':60 'round':1170,1192,1215,1312,1624,1949 'round-n.md':794 'rout':107,476,1804 'row':652,1955 'rule':698 'run':2,362,394,596,770,982,1016,1059,1119,1279,1298,1389,1393,1488,1632,1871,1959,1980,1989,2026,2036 'scan':892,940 'scope':905,1161,1210 'search':740,1757 'sebastianelvis/reaper':388 'second':224 'section':139,164 'secur':136,161,1105 'see':230 'select':1409 'self':1234,1474 'self-n':1473 'separ':526 'sequenti':1018 'server':247,252 'session':294 'set':267 'share':293,540 'shorter':618 'signal':545,1260,1358,1399 'signatur':128 'silent':256 'simpl':1156 'simpler':614 'simplic':612 'singl':561,719,1162 'skill':70,211,273,276,302,309,374,386,398,407,412,422,473,491,516,530,534,763,777,806,814,885,921,962,985,1505,1582,1627,1649,1668,1767,1776,1792,1795,1815,1821,1835 'skill-load':421 'skill-reaper' 'skill.md':358,427,469,493,1809 'skip':1067,1679 'slash':96,105,352,442,1591 'slash-command':95,104,441,1590 'sourc':562,810,823,1849 'source-sebastianelvis' 'spawn':995 'specif':837 'specifi':1701 'stage':434 'start':1621 'state':640,1367,1446,1831,1894 'status':855 'stay':586 'stdout':1756 'step':63,215,621,871,953,972,1081,1113,1185,1442,1463,1499,1514,1549,1551,1960,1981,1990,2007,2027,2037 'stop':593 'string':76,177,1589 'structur':555,629,1141 'stuck':833 'sub':308,397,411,515,1766 'sub-skil':307,396,410,514,1765 'subag':988 'substitut':345,511 'success':908 'supersed':1421 'support':241 'synthes':23,1501,2029 'synthesi':1308,1345,1519,1941 'tabl':649,1951 'tag':1457,1468 'take':48 'target':901,930,1633 'text':189 'threshold':127 'time':1619 'tip':645 'tool':1002,1793 'top':1681 'topic-agent-skills' 'topic-ai-research' 'topic-arxiv' 'topic-claude-code' 'topic-cline' 'topic-codex-cli' 'topic-cryptography' 'topic-cursor' 'topic-dblp' 'topic-distributed-systems' 'topic-gemini-cli' 'topic-iacr' 'total':1176,1199,1221 'track':1887 'trail':590,1545 'transit':329 'treat':193 'tri':569 'trigger':1402 'trust':1103 'truth':564,1851 'two':984 'typic':494 'unambigu':951 'uncertainti':599 'understand':865 'unresolv':1483,1920,1927,1997,2004,2015 'updat':655,667,1448,1647,1725,1732,1746 'usag':66 'use':26,87,288,341,399,416,819,989,1038,1770 'user':378,601,898,927,1534,1560,1610,1864 'v':1423 'vari':438 'venu':1762 'verdict':553 'wait':1601 'window':1875,1880 'windsurf':463 'without':111,262,957,1079 'work':100 'workflow':620 'workspac':537,624,627,632,844,860,1509,1525,1541,1825,1846,1859,1896,1907 'write':755,911,942,954,1020 'write-onc':754 'writer':720 'yet':870 'zero':785 'zero-pad':784","prices":[{"id":"ecd41f24-b2d7-4a5d-9f5c-276c950a4953","listingId":"4666c3fb-e4cf-4c3d-9c82-3ed0173a8ee4","amountUsd":"0","unit":"free","nativeCurrency":null,"nativeAmount":null,"chain":null,"payTo":null,"paymentMethod":"skill-free","isPrimary":true,"details":{"org":"SebastianElvis","category":"reaper","install_from":"skills.sh"},"createdAt":"2026-05-18T13:14:41.731Z"}],"sources":[{"listingId":"4666c3fb-e4cf-4c3d-9c82-3ed0173a8ee4","source":"github","sourceId":"SebastianElvis/reaper/reaper","sourceUrl":"https://github.com/SebastianElvis/reaper/tree/main/skills/reaper","isPrimary":false,"firstSeenAt":"2026-05-18T13:14:41.731Z","lastSeenAt":"2026-05-18T19:08:57.201Z"}],"details":{"listingId":"4666c3fb-e4cf-4c3d-9c82-3ed0173a8ee4","quickStartSnippet":null,"exampleRequest":null,"exampleResponse":null,"schema":null,"openapiUrl":null,"agentsTxtUrl":null,"citations":[],"useCases":[],"bestFor":[],"notFor":[],"kindDetails":{"org":"SebastianElvis","slug":"reaper","github":{"repo":"SebastianElvis/reaper","stars":8,"topics":["agent-skills","ai-research","arxiv","claude-code","cline","codex-cli","cryptography","cursor","dblp","distributed-systems","gemini-cli","iacr","openalex","research","scientific-research","semantic-scholar","skills"],"license":"apache-2.0","html_url":"https://github.com/SebastianElvis/reaper","pushed_at":"2026-05-02T06:06:07Z","description":"AI-native scientific research pipeline. Takes a research goal and autonomously runs multi-step academic research on Cursor, Codex CLI, Cline, Continue, Gemini CLI, Copilot, Windsurf, Claude Code, and 40+ more.","skill_md_sha":"1f59afaaef3775e15420dbc4a06b9b22b09c1569","skill_md_path":"skills/reaper/SKILL.md","default_branch":"main","skill_tree_url":"https://github.com/SebastianElvis/reaper/tree/main/skills/reaper"},"layout":"multi","source":"github","category":"reaper","frontmatter":{"name":"reaper","license":"Apache-2.0","description":"Run the full Reaper research pipeline: analyze a paper (optional), review literature, formalize the problem, investigate hypotheses with critique loops, and synthesize a report. Use when given a research goal, optionally with a paper.","compatibility":"Requires python3 with arxiv, requests, beautifulsoup4 packages, internet access, and PDF reading capability. The --codex flag requires an MCP-capable host."},"skills_sh_url":"https://skills.sh/SebastianElvis/reaper/reaper"},"updatedAt":"2026-05-18T19:08:57.201Z"}}