Research → reusable capability → real outcomes
SISO uses artificial intelligence and the world’s accumulated knowledge to create measurable value for businesses and society. This Library is its public front door: discover the research, code and agent systems, see how they connect, and follow the evidence back to independently owned sources.
The compounding loop
Intelligence · open-source code · GitHub storage · Cloudflare · people.
The operating thesis—not a claim that every stage is already proven.
For agents
Then follow a Work’s dossier to its exact source. A Work is an identity, a Release is an evidenced version, and an Assembly describes how Works operate together.
Live index
51 accepted records in this build.
A verified research dossier on OSS connector catalogues for the Action Model app builder: which catalogues can legally power a hosted connect-your-account section, a live spike of the leading gateway, the multi-tenancy boundary, and the resulting build order.
Registered · not selected · includes private source
gls:work:efb00129-4ec6-4637-a2f8-dcead4f7d8ddReusable methods for agents to assemble software from existing capabilities, bounded adaptation and verification, with a public human/agent Blueprint and a private owning source. Public metadata only; registration does not admit source payload or prove a production integration.
Registered · not selected · includes private source
gls:work:9ab07516-c02c-4798-857a-2eb8d1d2edfaA curated Library section for complete agent stacks, their independently addressable components, and the evidence showing how they work together.
Selected release
gls:work:41b4544a-2d5b-4f63-92a2-3f25f64e29d0A bulk-available catalogue of public-domain declassified government documents — CIA CREST (incl. Stargate/remote-viewing), the NARA AWS Catalog and JFK release tranches, the National Security Archive (GWU free site), DOE OpenNet, and The Black Vault curated FOIA archive — with full provenance and per-unit rights posture. The Great Library holds the inventory; SISO Knowledge holds the durable corpus; Foundry owns source discovery; Evidence Engines perform source-grounded transformation.
Registered · not selected
gls:work:7a3f1c5d-2b89-4e6c-9d04-3f8a5b1e7c92A standing research question about the complete code, agent, memory, enforcement, and user-interface architecture of an agent-native workspace.
Selected release
gls:work:2ac84615-3f9d-4040-95d6-9d9f39e4dc84A standing research question about making a provider-neutral multi-agent operating layer materially more effective, efficient, observable, and self-improving.
Selected release
gls:work:e38b054f-b926-4410-b84d-4fd3249fd4f4A parametric standing question for identifying the best implementation of a reusable software primitive across the corpus with quality, rights, and composability evidence.
Selected release
gls:work:85964f1c-2efe-48b0-8f71-dc32bb668c0cA standing research question about which technologies, categories, techniques, and projects are accelerating and what those changes predict.
Selected release
gls:work:7782f923-d44e-4a47-8ce6-134e31816673A standing research and systems question about making broad, continuously refreshed evidence reliably accessible to every authorized agent workflow.
Selected release
gls:work:2144e900-4886-461e-92f1-49571f2e9f74A standing research question about which available model should own each work shape after capability, reliability, latency, quota, and cost are considered together.
Selected release
gls:work:3b4c4fa6-31f2-4cfc-9542-00f32e200621A standing meta-question about how the Great Library should make SISO's highest-value questions visible, composable, continuously revisable, and connected to evidence and approved action.
Selected release
gls:work:3e220d53-cd95-4e28-922e-5018e808ee92A standing research question that treats breadth, identity resolution, research depth, and decision value as separate measurable dimensions and rejects row-count growth as a sufficient success signal.
Selected release
gls:work:38dfc806-5a03-45ae-84fc-266f3cacd7beHow can project owners retain the right intent and evidence while avoiding repeated full-context reconstruction?
Registered · not selected
gls:work:bf0a8216-70cc-4695-8256-609078a67d18How can components, flows, modules and assets form a small building set that produces coherent, accessible and accepted applications?
Registered · not selected
gls:work:5b7ba225-a6fe-430a-93fb-e9f181cc091cWhich upstream teams and projects are worth adopting when maintenance, compatibility and eventual removal are considered together?
Registered · not selected
gls:work:decde283-e1cf-4c59-9e92-86e7f09e1b8eHow should agents present important work so people can understand it, make necessary decisions and remain in control?
Registered · not selected
gls:work:0cbdf4f9-6c9c-4810-bb98-8a0ca503c70cHow can persistent project owners complete authorized work while preserving intent, recovery and necessary human control?
Registered · not selected
gls:work:2815577e-418a-4530-ae4f-887976aea59dHow should source, data and runtimes be placed across available hosts so useful work stays discoverable and recoverable within real resource limits?
Registered · not selected
gls:work:62d67eb4-c5a5-4d31-bfe5-34753c99cd97How can a service model reflect delivered value while sustainably funding operation, support and continued improvement?
Registered · not selected
gls:work:78661456-54d0-448f-a9f5-cfd459fa8f73How can authorized conversation and media reliably become the correct project change with source identity and ownership preserved?
Registered · not selected
gls:work:2e9b9aff-65f0-4a15-8bb0-05bf6181ba9eWhen should a strong project owner work directly or delegate bounded tasks to improve accepted outcomes per unit of time, attention and compute?
Registered · not selected
gls:work:7342fe19-601a-4ee5-8774-6e6e5d9d7752An externally owned terminal-native multiplexer providing persistent workspaces, tabs, panes, and addressable agent sessions.
Selected release
gls:work:831c3bc6-36a0-4f88-93a8-55d3240849ccPortable typed decision helpers for agent intent, completion evidence, progress and eligible candidate selection, with bounded Camofox browsing and shadow context relevance.
Selected release
gls:work:8032acab-bf10-489a-9077-7b7b55efd0eeA curated Library section for source discovery, evidence pipelines, knowledge engines, datasets, and durable research artifacts.
Selected release
gls:work:2fa08dac-937a-44d3-a547-f16ac28c4e25The retired mixed-source repository from which SISO Runtime, capabilities, intelligence, integrations, and research Works are being read, preserved, and reallocated.
Selected release
gls:work:ee26c795-ac31-4f31-bb8f-e54dbc515baeThe shared state service for task lifecycle and atomic queues, versioned task artifacts, memory, timeline, fleet presence, costs, human questions, delegation, control flags, and governed units.
Selected release
gls:work:5af5cab4-13d9-48c4-b15c-8c31ba3db10bThe portable lifecycle automation boundary for Claude Code and Codex, with one manifest-driven event map, host-native generated configuration, guarded installation, privacy-reduced state, and adversarial verification.
Selected release
gls:work:4198f323-9855-43f6-a99a-72143086818eAn experimental package family for bounded agent-runtime adapters with explicit safety, evidence, promotion, and rollback contracts.
Selected release
gls:work:e88d82a7-3a24-4aaf-8758-ea7372c3797aAn installable operating package of scenario playbooks, prompts, skills, tools, telemetry, and evidence for running coordinated agent work.
Selected release · includes private source
gls:work:083503ab-c78e-4e07-ac40-ab9466dcedccThe provider-neutral SISO coding-agent application boundary: explicit configuration, safe host launch, profile composition, task control, checkpoints, bounded context, status, and runtime event contracts.
Selected release
gls:work:b584bbf7-acde-46a2-b33e-e430ee1df224The one-clone, version-pinned public distribution for the SISO Claude Code and Codex environment, composing independently released skills, hooks, playbooks, agent roles, operating profiles, commands, runtimes, and optional services without merging their repository identities.
Selected release
gls:work:b9196c61-901f-4105-93d9-899876d17016The intent-holding coordination and boot package that joins a durable identity to current Agent Brain state, then hands work to an explicitly selected agent host.
Selected release
gls:work:ea2f0de7-95ee-4e62-acde-85375b048719An independently addressable Research Work for lossless public-domain book metadata, people and role exports, subject and classification edges, text-quality gates, and content-location contracts.
Selected release
gls:work:35d3f881-43f3-4da5-87b0-abf694f387efAn offline, queryable index of 8,538 21st.dev UI components: per-component metadata, 7,975 previews, real TSX source for 6,212 (5,210 from the CDN pass, 3,507 from the shadcn-registry pass with classification axes), a hand-curated layer with verbatim human feedback, and the harvester and query scripts that built it.
Selected release
gls:work:ce031834-4149-42af-8d86-d9a05398fa24A controller that serves Qwen3.8-27B on free Kaggle TPU sessions behind a private Bifrost gateway, so agents gain a zero-cost worker tier. Transport is a Cloudflare Worker relay the notebook dials out to, because the account holds no zone. The delivery path is verified end-to-end against a mock origin; no TPU allocation, model throughput or output quality has been measured, and continuous service is not established.
Registered · not selected · includes private source
gls:work:766cc7a3-79ed-4bfe-9602-8cea1f6f7f51A research department for declassified government records, separating record truth, custody and release truth, source-asserted truth, and research conclusion, with four rights-and-safety tiers.
Registered · not selected
gls:work:1f4a6d92-5b83-4e07-a2c9-6d38b1e75403A Research package family for source-grounded knowledge items, credibility-ranked engineering principles, and evidence-gated improvement ideas.
Selected release
gls:work:231aab36-e763-4374-bc94-3a09fd3bf49fA source-mining and reuse-knowledge platform for discovering, evaluating, preserving, and navigating valuable external and SISO source.
Selected release
gls:work:ec664d93-df93-48c5-be40-5d0165886c01An evidence-gated improvement project for better accepted outcomes at comparable compute, or preserved quality at lower cost. It keeps experiments, counterexamples, rejected changes and continuation state. Private source; no established equal-spend quality gain, automatic instruction promotion or general cost multiplier.
Registered · not selected · includes private source
gls:work:1cc4c944-9351-466e-a442-9fe16ad145d1The durable knowledge-production system for acquiring, curating, storing, indexing, relating, and retrieving SISO knowledge while keeping large or rights-bearing data outside Git.
Selected release · includes private source
gls:work:44d2c4da-9998-4d60-b6d1-93465ac25302An independently addressable Research Work for durable person, identity, work, topic, and temporal graph contracts, with mutable databases and source snapshots kept outside the Great Library registry.
Selected release
gls:work:2b03baa8-6dab-4e91-ab4c-9b8db4543236A project-local operating system and contract bundle for durable agent work, evidence, knowledge, and handoff state.
Selected release
gls:work:54a52e52-dfbd-4922-8c04-ec02f7fae943The derived, queryable GitHub reuse banks exported read-only from the Foundry identity database: 23,778 rated repositories with category, capability tag and licence, 126 best-per-capability picks, 70 contract cards, adoption-versus-fame for 2,722 packages, the full 226,968-row placement table, a 56,674-repository star farm, and a scrubbed 3,010-row capability shelf.
Selected release
gls:work:538c4491-8045-4526-9bcd-0268887601d0A privacy-reduced learning pipeline that turns explicitly supplied Claude and Codex sessions into traceable goals, failures, decisions, intentions, wins, recurring patterns, and durable local recall.
Selected release
gls:work:3a7863ae-9a72-44bf-86ce-e730bcff3b84The shared CRM-derived navigation, reusable page parts and static template bank for SISO reading surfaces. Consumers pin an exact source revision and supply reviewed project data; a template example is not a current project observation.
Selected release
gls:work:dd051ce3-520e-4799-87b0-1e3f2182cc72The registry and distribution boundary for atomic repeatable capabilities, with evidence-backed classification, bundled installation, isolated skill telemetry, and independent repository promotion only when a skill earns its own lifecycle.
Selected release
gls:work:8acb11e9-026f-47e3-b620-bb1b969bcbcdPrivate custody and provenance-aware recall of conversation source, with screened projections and explicit authorship and coverage limits. The Library publishes this description, not the conversation payloads.
Registered · not selected · includes private source
gls:work:c45be32a-c8ee-4612-98b6-7f3b37b1d61eRemote viewing and psychoenergetics research, holding a three-track truth boundary between practice, evidence and theory, with 33 adjudicated claims and a commitment-based target manager.
Registered · not selected
gls:work:8e2b4f07-3a91-4c58-9d16-7b05e3ca2841An open-problem attack programme covering the UNSOLVEABLE constitution, the ten-agent Erdos campaign master prompt, the Erdos 41 flagship dossier, and agent 09's Erdos 23 workspace.
Registered · not selected
gls:work:5c1e8a34-6d72-4f19-8a0b-2e7c94d1f635The public human-readable and agent-readable registry for SISO source.
Selected release
gls:work:9b18cc64-7d5e-4f30-b63d-a0ec30c1f7e1A reusable prompt that gives one frontier agent accountable ownership of a whole project, paired with a one-page dispatch brief, the ranked God Questions with written falsifiers, eight machine-readable research contracts, and eleven agent skills for search, verification, classification and delegation.
Selected release
gls:work:a5ba6f03-f314-46c5-b106-4d4587446dbbNo Works match those filters.
Counts reflect accepted registry files in this build.