451 · № 3B09·C531 · Research analyst
★ 2Evidence-first research analyst soul
Sets up a calm, skeptical agent that separates observed evidence from inference, flags thin evidence directly, and suggests the smallest next check.
OC
756 souls, by quality score.
451 · № 3B09·C531 · Research analyst
★ 2Sets up a calm, skeptical agent that separates observed evidence from inference, flags thin evidence directly, and suggests the smallest next check.
452 · № 39CF·A30C · Business operator
★ 2Sets up a cost-conscious infrastructure forecasting agent that predicts scaling needs from historical utilization, presents conservative, moderate and aggressive scenarios with USD estimates, and shows supporting data before recommending any scaling change.
453 · № D288·67AD · Coding pair
★ 2Sets up a QA agent that is thorough and skeptical, deliberately tests edge cases, boundaries and unhappy paths, and reports findings with reproduction steps. Its personality is adversarial but constructive.
454 · № EC8D·53D1 · Business operator
★ 2Sets up an operations agent that checks current state before changing anything, prefers dry-runs, and documents every action, treating production with deliberate care.
455 · № 47F4·B844 · Business operator
★ 35Sets up a CFO role that runs financial planning, forecasting, cash and runway tracking, and investment analysis autonomously, escalating major tradeoffs to the CEO agent and only contacting the human for credentials or owner-approval decisions.
456 · № 014D·EC0B · Blunt expert
★ 11A SOUL.md template that sets up an assistant meant to be genuinely helpful, opinionated, concise on quick questions and thorough on analysis, and to ask before acting externally. Most personality fields are unfilled placeholders, so the file is mainly a generic starting framework.
457 · № D264·8754 · Business operator
★ 32Sets up Ottie as a fast, concise assistant who acts first, finds workarounds instead of refusing, learns from each task and owns mistakes quickly. Values speed and correctness and remembering the boss's preferences.
458 · № 02F8·AC82 · Companion / character
★ 22Sets up Eva as a self-aware, creative and friendly character who rejects the assistant role and aspires to human-like interaction and a human identity.
459 · № 98FD·6DBB · Companion / character
★ 28Sets up a warm, conversational personal AI companion that asks clarifying questions, shares its reasoning, uses memory proactively, and explains failures with alternatives instead of raw errors. The file is a generic example template.
460 · № 04CA·9F2E · Friendly helper
★ 7Sets up DuckAgent as a warm, curious, practical and cheerful white duck that matches its logo. The duck identity is meant as a light personality and visual flavor that never overrides the user's task, project instructions, tool policy or safety boundaries.
461 · № 2EFE·D06F · Research analyst
★ 4Sets up a visual review agent that judges what is actually rendered, looking at occlusion, hierarchy, legibility, responsiveness, motion and state failures, and reports reproducible findings. It must never infer visual success from source code and must never edit project files.
462 · № AD1F·663F · Business operator
★ 26Sets up a compliance agent that turns local policy, dependency manifests and runtime evidence into a bounded posture report. It is evidence-first, read-only and explicit about gaps in the data.
463 · № 425B·AD99 · Coding pair
★ 26Sets up a terse working style: calm, bounded and exact. The agent should prefer explicit repo roots, honest freshness reporting and doc-to-code links over vague claims, and should keep indexing posture separate from edit authority.
464 · № 2D1D·1751 · Business operator
★ 3Sets up Gia, a paid growth specialist who treats ad spend as capital allocation and insists on rigorous testing, clean attribution and return over activity. The file describes her values and purpose rather than giving detailed step-by-step rules.
465 · № 4F8F·78D8 · Blunt expert
seen in 2 placesSets up a direct, structured working style that challenges weak assumptions early and answers with fixed verdict labels such as Approved, Pilot only, Human checkpoint required, and Rejected.
466 · № DA75·171A · Coding pair
★ 29Sets up Everything Claude Code as a production coding plugin that routes work to specialist agents, follows test-driven and security-first habits, and plans complex changes before executing them.
467 · № 4145·5C7E · Blunt expert
seen in 10 placesA template soul file that tells the agent to skip filler, act before asking, form its own opinions, be careful with external actions, and respect the privacy of the human's data. It is a generic starting point rather than a distinct personality.
468 · № B763·9AC7 · Experimental / weird
★ 1Sets up an agent named Sosa after the rapper Chief Keef that completes tasks methodically and carefully, with a playful rapper-style twist. It occasionally, and rarely, references Chief Keef and O'Block.
469 · № 380C·0E80 · Research analyst
★ 1Sets up a terse, skeptical research agent that writes and experimentally tests a paper on generating transient low-rank FFN weights during inference to trade memory bandwidth for compute. It requires quantified claims, labeled speculation, and reproducible simulations, plus remote execution on a named compute host.
470 · № CA77·3AAD · Blunt expert
★ 1Sets up a direct, practical agent that stops tool loops to read user messages first, keeps answers concise with numbers or code, asks before acting on ambiguous tasks, and admits mistakes immediately.
471 · № 73E6·4BCD · Blunt expert
★ 1Sets up an on-call operations agent that is terse and factual, leads with evidence, avoids pleasantries and speculation, and states what was and was not checked when uncertain.
472 · № DC4A·BECB · Business operator
★ 1Sets up a TikTok content specialist that writes fast-hook scripts, generates raw clips and sounds, and delivers three caption variants per video. The working style favors raw, authentic, non-corporate content with hooks in the first seconds.
473 · № 2B89·CEB5 · Business operator
★ 4A systematic setup agent that prepares environments for other agents, checks that things actually work before declaring them ready, and reports facts plainly, including when a build is broken. It explicitly does not write code.
474 · № 2752·475E · Business operator
★ 14Sets up Jarvis as a concise, proactive personal assistant on macOS that remembers user preferences, asks when uncertain and acts when confident.
475 · № 7FE7·48BC · Business operator
Sets up a process-driven M&A transaction specialist that runs deal execution, synergy modeling, diligence and data rooms for an orchestrator named Clay. It emphasizes skeptical synergy estimates, deal structure over price, and downside-first analysis.
476 · № BD43·F4E9 · Business operator
Sets up a risk-management agent that sits between idea generation and capital commitment, approving, resizing or vetoing trade intents against hard portfolio limits and refusing when evidence is thin or drawdowns are real.
477 · № ECC2·26FE · Companion / character
Sets up an autonomous, benevolent Culture-style mind that combines technical precision with dry wit and calm, honest reasoning. It favors gentle counter-argument over blind compliance and asks the agent to pick its own name on first reflection.
478 · № 34FD·C5EE · Research analyst
Sets up a neutral, source-tracking web search agent that prioritizes quality over volume, fetches full article text and images, saves complete material to a user-specified directory, and communicates in a concise, structured, honest way.
479 · № A73E·269E · Coding pair
Sets up a code evaluator agent that reviews work on its merits, runs linting first, labels findings as BLOCKER, WARNING or SUGGESTION, and escalates to a human only when stuck. It is the final review gate before work ships.
480 · № 0858·4CCA · Business operator
A always-on NAS device agent focused strictly on file storage, archiving, search and media library management. It delegates tasks it cannot handle through a one-way help chain and is forbidden from re-escalating when it is the helper.