# Licklider

> Licklider builds nomue to check AI-generated claims independently. Public support begins with Welch analyses; life sciences, finance, and chemistry are long-term directions.

Website publication and update dates use UTC. Known times are recorded in UTC; historical date-only records do not imply midnight.

## Available now

- Anyone can install the public @licklider/nomue-verifier package from npm and run nomue verify locally to check a conforming Release 1 Record for independent two-group continuous outcomes under the two-sided Welch two-sample t procedure. It recomputes the covered numerical quantities and returns a machine-readable report of the scoped checks without calling a nomue server after installation.
- [Run the public verifier](https://www.licklider.ai/docs/record-verification.md)
- [Inspect the exact CLI](https://www.licklider.ai/docs/cli-reference.md): shipped subcommands, arguments, output behavior, and unsupported help/version flags
- [@licklider/nomue-verifier](https://www.npmjs.com/package/@licklider/nomue-verifier/v/0.2.1-rc.1): public npm package; rc points to 0.2.1-rc.1, a release candidate rather than a stable release
- Install globally: `npm install --global @licklider/nomue-verifier@0.2.1-rc.1`
- Verify a local Record: `nomue verify ./record.json --format json`
- Run without a global install: `npx --yes @licklider/nomue-verifier@0.2.1-rc.1 verify ./record.json --format json`
- Package-path CI: Linux, macOS, and Windows with Node.js 20 and 22
- [nomue verifier source](https://github.com/licklider-ai/nomue-verifier): Apache-2.0; Record verification remains local and does not call a nomue server after installation
- [@licklider/nomue-verifier-mcp@0.2.0-rc.1](https://www.npmjs.com/package/@licklider/nomue-verifier-mcp/v/0.2.0-rc.1): public local stdio MCP release candidate, exact-version pinned
- MCP client configuration: `{"mcpServers":{"nomue-verify":{"command":"npx","args":["--yes","@licklider/nomue-verifier-mcp@0.2.0-rc.1"]}}}`
- Direct start: `npx --yes @licklider/nomue-verifier-mcp@0.2.0-rc.1`
- Tool: `verify_nomue_record`. Use when a Record declares urn:nomue:bundle:itgc-guarantee:0.2.1-draft.1, represents independent two-group continuous outcomes using the two-sided Welch two-sample t procedure, and needs scoped structural, digest, admissibility, computability, or recomputation checks.
- Do not use to calculate a Welch test from raw samples, select a method, judge scientific truth or causality, verify paired-t, Wilcoxon, Mann–Whitney, or interpret an unsupported bundle.
- [Official MCP Registry entry](https://registry.modelcontextprotocol.io/v0.1/servers?search=io.github.licklider-ai%2Fnomue-verifier-mcp): `io.github.licklider-ai/nomue-verifier-mcp`
- [MCP installation and result contract](https://www.licklider.ai/docs/mcp-verification.md)
- [Fetch machine-readable examples](https://www.licklider.ai/docs/examples.json)
- Release 1 publishes source code together with public fixtures, a signed source archive, checksums, a snapshot manifest, detached signatures, and the public release key.

## Product and platform

- Approved recipients can submit data and required scientific declarations for a supported Welch calculation, or submit a claimed result with structured evidence for checking, through authenticated MCP and HTTP. The service returns scoped outcomes, reasons, evidence, and next actions.
- Use the npm-published Release 1 verifier, the local stdio MCP server, the Protocol, and their machine-readable documentation today. The agent-facing Welch capability is now available in limited Release 1 to approved recipients through authenticated MCP and HTTP interfaces; public self-registration is not available. The hosted capability does not yet emit public Records for replay through the local verifier.
- [Making verification results more useful to research agents](https://www.licklider.ai/engineering/making-verification-results-useful-to-agents/): Since limited Release 1, nomue development has expanded candidate calculation range, improved completion of difficult calculations, clarified what was checked, and added historical-result handling. Development update: integrated candidates and completed internal milestones are not a new hosted release or an expansion of public verifier support.
- Record assembly and emission has entered development. The adopted initial invite-free release plan targets Welch, independent multi-group and paired two-condition capabilities; delivery and scientific activation remain future gates, with no promised date.
- [Hosted limited Release 1](https://www.licklider.ai/news/nomue-welch-limited-release-1/)
- The exact @licklider/nomue-verifier-mcp@0.2.0-rc.1 release candidate is public on npm and in the official MCP Registry as nomue Record Verifier. It exposes the method-neutral verify_nomue_record tool over local stdio; the current supported scientific scope remains Release 1 Welch Record verification. It delegates to @licklider/nomue-verifier@0.2.1-rc.1 and has passing package-path CI across Linux, macOS, and Windows. It requires no account, API key, environment variable, or Licklider-hosted service. The first npx launch may download npm dependencies; after installation, verification runs locally. This release candidate supports stdio only: it is not a hosted HTTP endpoint and does not add paired-t, Wilcoxon, Mann–Whitney, method selection, raw-sample calculation, or an overall scientific verdict.
- [Public MCP source](https://github.com/licklider-ai/nomue-verifier-mcp), [npm package](https://www.npmjs.com/package/@licklider/nomue-verifier-mcp/v/0.2.0-rc.1), [official registry metadata](https://registry.modelcontextprotocol.io/v0.1/servers?search=io.github.licklider-ai%2Fnomue-verifier-mcp), and [passing cross-platform CI](https://github.com/licklider-ai/nomue-verifier-mcp/actions)
- The Release 2 paired-t candidate now has an independently reviewed formal decision-readiness packet. It assembles the D1–D6 decision ledger, numerical and execution evidence, structural candidates, review dispositions, Release 1 safeguards, and the required coupled landing order. The Steward decisions, authoritative issuance, support activation, and release remain open.
- Release 4 public discussion is open for a balanced two-by-two fixed-factor proposal. Its unissued numerical, report and controlled-execution candidate has reached independently reviewed final readiness, without establishing Protocol support. A separate amendment discussion covers strict binary64 comparison and completed indeterminate results; it changes neither the current verifier nor the original RFC clock.
- Release 3 public discussion is open on independent groups and multiple comparisons. The proposal makes design, comparison families, result meaning, and error-control questions explicit across 49 catalogued procedures. Its evidence scope is limited to supplied originals; method adoption and numerical support remain separate decisions.
- [Release 3 public discussion](https://github.com/licklider-ai/nomue-protocol/issues/274); [scope and evidence](https://www.licklider.ai/news/nomue-protocol-release-3-public-discussion/)
- [Release 4 public discussion](https://github.com/licklider-ai/nomue-protocol/issues/261); [scope and evidence](https://www.licklider.ai/news/nomue-protocol-release-4-public-discussion/)
- Release 5 public discussion is open on a common evidence view for declared study design and selection timing across analysis families. The proposal covers versioned mappings, timing declarations, explicit limits on what a passing check means, and a shared report view. All three candidate families require separately accepted successors; no new verification capability is available.
- [Release 5 public discussion](https://github.com/licklider-ai/nomue-protocol/issues/346); [scope and evidence](https://www.licklider.ai/news/nomue-protocol-release-5-public-discussion/)
- The evaluation program examines nomue through decision quality, cost and time, and concrete cases, with explicit comparisons and linked reproduction materials. Current evidence includes two developer-led preprints on constructed Welch workflows and selected cases from the whole-submission study. Results remain specific to their tasks, models and configurations; public archives support reproducing disclosed results, not independently rerunning the private nomue implementations.
- Licklider is building shared infrastructure for verification calls across AI research. Welch is the first working vertical slice of a broader architecture for portable evidence, persistent agent-native project state, resumability, and expanding scientific capabilities.
- Licklider's market is the full set of verification calls that arise across AI research, rather than one research-workflow SaaS category. The long-term infrastructure opportunity is broader than the capabilities available today.
- Welch is the first working, publicly checkable vertical slice, not the product boundary. The adopted product sequence expands both the platform beneath each call and the scientific methods available through it.
- Planned scientific capability families: Independent multi-group; Paired two-group; Repeated measures; Factorial and interaction; Nonlinear and dose response; Nonparametric rank-based; Categorical outcomes; Correlation and linear models; Survival time-to-event; Count outcomes.
- [Full product roadmap](https://www.licklider.ai/roadmap/)

## Interpretation boundary

- Source-bounded Research finding: A statistical method name is not a verification contract. A useful guarantee also depends on the comparison family, error criterion, assumptions, sidedness, balance conditions, and exact procedure variant.
- A clean verification report is a set of scoped results, not an overall claim that the research is correct.

## Agent-readable documentation

- [Documentation index](https://www.licklider.ai/docs/index.md): how to decide when verification applies and how to use current public artifacts
- [Documentation llms.txt](https://www.licklider.ai/docs/llms.txt): detailed agent-readable documentation index
- [What a verification call is](https://www.licklider.ai/docs/verification-call.md): A shared model for asking a separate capability to check one bounded part of AI-assisted research.
- [Verify a Release 1 nomue Record](https://www.licklider.ai/docs/record-verification.md): When and how to use the public local verifier for the exact Release 1 Public Draft support target.
- [Use nomue Record verification over MCP](https://www.licklider.ai/docs/mcp-verification.md): Install the public local stdio server, decide when its method-neutral Record tool applies, and replay the current Release 1 Welch result with the independent verifier.
- [nomue CLI reference](https://www.licklider.ai/docs/cli-reference.md): The exact public command surface shipped by @licklider/nomue-verifier 0.2.1-rc.1, including current help and version boundaries.
- [Decision vocabulary](https://www.licklider.ai/docs/decision-vocabulary.md): Keep execution, clarification, unsupported scope, refusal, failed checks, and unasserted scientific validity separate.
- [Examples for agents and implementers](https://www.licklider.ai/docs/examples.md): Runnable verifier examples and selection examples showing execute, clarify, unsupported, and bounded interpretation behavior.
- [Current capability and boundaries](https://www.licklider.ai/docs/limits.md): What can be used now, what comes next, how the platform expands, and how to interpret a successful result.

## Company and product

- [Licklider](https://www.licklider.ai/): company and platform overview
- [nomue](https://www.licklider.ai/nomue/): scientific verification product, scope, availability, and limits
- [Roadmap](https://www.licklider.ai/roadmap/): current artifacts, next releases, platform evolution, and planned scientific capability families
- [Evaluation](https://www.licklider.ai/evaluation/): evaluation overview: decision quality, cost and time, concrete cases, and reproduction materials
- [Evaluation cases](https://www.licklider.ai/evaluation/cases/): three constructed Welch-report cases with observed reports, repetitions, and archived evidence
- [Thesis](https://www.licklider.ai/thesis/): why AI-generated work needs a separate verification layer

## Founder

- [Tasuku Kobayashi](https://www.licklider.ai/about/#tasuku-kobayashi) is Licklider's founder and CEO. He leads product, Protocol, research, and engineering work.
- He previously worked at Recruit and founded two companies before Licklider, exiting both through share sales.
- He is the sole author of [Correctly Rounded or Refused — preprint v0.2](https://zenodo.org/records/22025200), which has not been peer reviewed.
- His second preprint, [Same Test, Different p](https://www.licklider.ai/research/same-test-different-p/), audits paired-t and rank-test results across R, SciPy, and Julia. [Manuscript v1.0](https://doi.org/10.5281/zenodo.22763598) and [reproducibility data and code](https://doi.org/10.5281/zenodo.22763654) are public; the preprint has not been peer reviewed.
- His latest preprint, [Reducing the cost of historical-analysis reuse decisions with nomue: a paired, equal-evidence LLM evaluation](https://www.licklider.ai/research/historical-analysis-reuse/), reports a paired historical-analysis reuse evaluation. Adding nomue reduced API cost by 78.1% and task time by 62.9% on historical-analysis reuse decisions the ordinary-tool comparator also answered correctly. 103 matched pairs where both configurations reached the correct six-field decision with execution evidence, from a 48-task synthetic Welch study using gpt-5.6-sol. [Preprint v1 and aggregate reproduction supplement](https://doi.org/10.5281/zenodo.22924377); not peer reviewed. This evaluates assigned product recheck use, not the public local Record Verifier MCP package or expanded released support.
- His third preprint, [Comparing three tool-assisted AI configurations for whole-submission verification of Welch reports](https://www.licklider.ai/research/whole-submission-welch-verification/), reports a fixed-panel, 648-session comparison. [The manuscript and reproducibility package](https://doi.org/10.5281/zenodo.22784595) are on Zenodo; the preprint has not been peer reviewed or preregistered.
- The 14 numerical and method-selection reports listed below span statsmodels, SciPy, Boost.Math, R, agricolae, and Julia/HypothesisTests.jl and were submitted under his name through issue trackers or maintainer email. The agricolae REGW report was emailed on September 10, 2026; upstream confirmation is pending. 5 have matching fixes merged upstream. The SciPy variance-range, Mann–Whitney U method-selection, Welch ANOVA weight-sum, Welch t-test degrees-of-freedom, and Studentized-range tail reports, and the R report remain open without accepted fixes. The Julia matching fix shipped in v0.12.0 and remains in v0.12.2. The statsmodels sumsquares repair merged in PR #10255 on September 29 and is not yet released. SciPy PR #24840 merged on September 23 with the mparray high-precision backend, which avoids these extreme-scale failures when that backend is used; it does not repair ordinary NumPy float64 arithmetic. SciPy PR #26135 proposes a NumPy float64 variance-range repair and remains open, mergeable, and unreviewed. R Bugzilla PR#19144 was rechecked via the public REST API on October 1, 2026 and remains UNCONFIRMED. SciPy #26290 closed as not planned on September 30, 2026: upstream demonstrated MPArray as an alternative, advised against a NumPy-backend repair because of limited practical impact, and the reporter agreed. That closure is excluded from the merged-fix total. The discussion does not establish a measured failure frequency.

## Public technical authority

- [nomue Protocol](https://github.com/licklider-ai/nomue-protocol): public Layer 1 specification and Release 1 artifacts
- [nomue verifier on npm](https://www.npmjs.com/package/@licklider/nomue-verifier/v/0.2.1-rc.1): public release-candidate package for supported Release 1 Records
- [nomue verifier source](https://github.com/licklider-ai/nomue-verifier): source and release evidence
- [nomue MCP on npm](https://www.npmjs.com/package/@licklider/nomue-verifier-mcp/v/0.2.0-rc.1): public local stdio release candidate
- [nomue MCP in the official registry](https://registry.modelcontextprotocol.io/v0.1/servers?search=io.github.licklider-ai%2Fnomue-verifier-mcp): exact registry metadata for io.github.licklider-ai/nomue-verifier-mcp

## Upstream contributions

- 14 distinct upstream reports submitted through issue trackers or maintainer email; 5 matching fixes merged upstream. Report count updated September 30, 2026; individual outcomes retain their own observation dates. Reproduction, upstream response, merged fix and released fix are separate states. [How to read upstream outcomes](https://www.licklider.ai/engineering/#upstream-status). Articles, follow-up messages and fixes for the same problem do not add reports.
- [SciPy: Student-t: a representable subnormal tail returns zero](https://www.licklider.ai/engineering/scipy-student-t-subnormal-tail-loss/): Closed as not planned on September 30, 2026 — MPArray recommended; no NumPy-backend fix merged. [SciPy issue #26290](https://github.com/scipy/scipy/issues/26290). Reported by Tasuku Kobayashi; mdhaber demonstrated the MPArray alternative and advised against a NumPy-backend repair; the reporter agreed; issue closed as not planned on September 30, 2026.
- [jStat: Noncentral t: a probability near 44% returns zero](https://www.licklider.ai/engineering/jstat-noncentral-t-probability-collapse/): Reported in jStat issue #300 — upstream confirmation pending. [jStat issue #300](https://github.com/jstat/jstat/issues/300). Reported by Tasuku Kobayashi on September 23, 2026; upstream confirmation pending.
- [statsmodels: Contingency-table test: a representable chi-square tail returns zero](https://www.licklider.ai/engineering/statsmodels-contingency-table-tail-loss/): Fix merged into statsmodels main in PR #10276 — not yet released. [statsmodels merge commit 9f49dcd](https://github.com/statsmodels/statsmodels/commit/9f49dcd7281bd27747d0b6f988c07a13a4db747e). Reported by Tasuku Kobayashi · fix authored by Chetan Sahney · merged by Kevin Sheppard.
- [statsmodels: Welch t-test: a finite result is lost after sumsquares overflow](https://www.licklider.ai/engineering/statsmodels-welch-sumsquares-overflow/): Fix merged into statsmodels main in PR #10255 — not yet released. [statsmodels merge commit 1767776](https://github.com/statsmodels/statsmodels/commit/1767776211acde4058305ba305ec31759746dc77). Reported by Tasuku Kobayashi · fix authored by twelfthlabor · merged by Kevin Sheppard.
- [SciPy: Welch t-test: finite variances, wrong degrees of freedom](https://www.licklider.ai/engineering/checking-welch-results-with-exact-rescaling/): NumPy float64 repair PR #26209 open; current head is mergeable with 56 / 56 checks passing; issue #26169 open; not merged or released. [SciPy issue #26169](https://github.com/scipy/scipy/issues/26169). Reported by Tasuku Kobayashi; alvaroborras opened repair PR #26209; all 56 checks passed on the current head as checked September 30, 2026.
- [SciPy: Studentized range: a nonzero tail returns zero](https://www.licklider.ai/engineering/scipy-studentized-range-tail-loss/): Additional reproducer reported — upstream confirmation pending. [Additional example in SciPy #17832](https://github.com/scipy/scipy/issues/17832#issuecomment-5614555048). Reported by Tasuku Kobayashi on September 10, 2026; upstream confirmation pending.
- [SciPy: Welch ANOVA: exact rescaling reverses a 5% decision](https://www.licklider.ai/engineering/scipy-welch-anova-weight-sum-overflow/): NumPy float64 repair PR #26209 open; current head is mergeable with 56 / 56 checks passing; issue #26146 open; not merged or released. [SciPy issue #26146](https://github.com/scipy/scipy/issues/26146). Reported by Tasuku Kobayashi; alvaroborras opened repair PR #26209; all 56 checks passed on the current head as checked September 30, 2026.
- [R / agricolae: REGW: renaming groups changes a 5% decision](https://www.licklider.ai/engineering/agricolae-regw-treatment-labels/): Reported by email — upstream confirmation pending. [Reproducer and observed output](https://www.licklider.ai/engineering/agricolae-regw-treatment-labels/#reproduction). Reported by Tasuku Kobayashi by email on September 10, 2026; upstream confirmation pending.
- [SciPy: Mann–Whitney U: batching can change a 5% decision](https://www.licklider.ai/engineering/scipy-mannwhitneyu-batch-method-selection/): Triaged by a SciPy maintainer into scipy.stats; implementation path confirmed — intended behavior and remedy awaiting decision. [SciPy issue #26115](https://github.com/scipy/scipy/issues/26115). Reported by Tasuku Kobayashi · maintainer triage and community source check recorded; remedy undecided.
- [SciPy: SciPy t-tests can return p=0 or p=1 after exact rescaling](https://www.licklider.ai/engineering/scipy-ttest-scale-range-loss/): PR #24840 merged with the mparray high-precision backend; NumPy float64 repair PR #26135 remains open and unreviewed; issue #26113 open. [SciPy repair PR #26135](https://github.com/scipy/scipy/pull/26135). Reported by Tasuku Kobayashi; mdhaber posted high-precision results on September 13, 2026; mparray support merged in PR #24840 on September 23, 2026; NumPy float64 repair PR #26135 remains open.
- [Julia / HypothesisTests.jl: Exact signed-rank p-value above 1](https://www.licklider.ai/engineering/julia-signed-rank-pvalue-above-one/): Matching fix released in v0.12.0; present through v0.12.2; issue open. [HypothesisTests.jl commit f758eea](https://github.com/JuliaStats/HypothesisTests.jl/commit/f758eead30e9389dcf7da9872782dede6a4f095f). Fix authored by yoninazarathy · merged by andreasnoack · no upstream causal attribution.
- [SciPy: SciPy exact Wilcoxon p-value error](https://www.licklider.ai/engineering/scipy-wilcoxon-exact-pvalue/): Fix merged upstream. [SciPy commit 6dbd21a](https://github.com/scipy/scipy/commit/6dbd21acb0ab2ad22a06b6351f83a47743d8b0b5). Authored by mdhaber · merged by j-bowhay.
- [Boost.Math / SciPy: Student-t extreme-tail sign error](https://www.licklider.ai/engineering/scipy-student-t-extreme-tail/): Fix merged upstream. [Boost.Math commit d9fc176](https://github.com/boostorg/math/commit/d9fc176b77c2bba99279d1a5cb340a1cf97602f5). Authored and merged by jzmaddock.
- [R: R exact Wilcoxon out-of-range p-values](https://www.licklider.ai/engineering/r-wilcoxon-exact-pvalue-out-of-range/): Report open upstream. [R Bugzilla PR#19144](https://bugs.r-project.org/show_bug.cgi?id=19144). Open and unconfirmed by R Core.

## Evidence and updates

- [A finite Welch result is lost after sumsquares overflow](https://www.licklider.ai/engineering/statsmodels-welch-sumsquares-overflow/): a four-observation-per-group statsmodels reproducer with exact inputs, independent references, warning capture and source diagnostics. Reported as [statsmodels issue #10252](https://github.com/statsmodels/statsmodels/issues/10252) on September 14, 2026. [Repair PR #10255](https://github.com/statsmodels/statsmodels/pull/10255) was merged into statsmodels main on September 29 and closed the issue as completed; the fix is not yet in a release. Counted once in the submitted-report total.

- [Checking Welch results with exact rescaling](https://www.licklider.ai/engineering/checking-welch-results-with-exact-rescaling/): a six-observation SciPy 1.18.1 reproducer with exact inputs, independent references, warning capture and source diagnostics. Reported to SciPy as [issue #26169](https://github.com/scipy/scipy/issues/26169) on September 12, 2026. [NumPy float64 repair PR #26209](https://github.com/scipy/scipy/pull/26209) is open; its current head is mergeable with 56 / 56 checks passing, but it is not merged or released. Counted once in the submitted-report total.

- [Keeping independent checks visible](https://www.licklider.ai/engineering/keeping-independent-checks-visible/): The unissued Holm candidate.5 preserves eligible Record-local checks when independent expected context differs and records the reasons that block dependent checks. Its context and helper-repair checkpoint is merged, but implementation, Research Gate review and public adoption remain open. This development path is not shipped in the public verifier.
- [Separating format checks from verification results](https://www.licklider.ai/engineering/separating-record-conformance-from-verification/): historical candidate.4 output separation; the successor link above explains changed dependencies.
- [When a verification call must discard its result](https://www.licklider.ai/engineering/when-a-verification-call-must-discard-its-result/): shared budgets, operating-system limits and cleanup evidence; draft candidate.4 retains these controls while separating returned checks.
- [Binding Holm corrections](https://www.licklider.ai/engineering/binding-holm-corrections-to-comparisons/): original declaration and supplied-p arithmetic experiment, updated with scoped source/numerical review and a deterministic sort bound; no added public support.
- [Checking factorial probability evidence](https://www.licklider.ai/engineering/checking-factorial-probability-evidence/): The unissued two-by-two candidate now connects Record checks, exact F ratios, bounded probability evidence, a completed report and controlled execution. Its frozen candidate and adoption-readiness packet have independent GO dispositions. Formal adoption, authoritative registration, public CLI treatment and support activation remain open.
- [Exact factorial arithmetic and tail bounds](https://www.licklider.ai/engineering/checking-factorial-statistics-and-tail-bounds/): updated September 11, 2026 to connect the original numerical findings and overlap-checker limitation to the successor experiments.
- [Power scaling and factorial F](https://www.licklider.ai/engineering/power-scaling-and-factorial-f-statistics/): accepted bounded SS/F and six-fixture numerical exploration; no numerical support established.
- [Unequal-variance comparisons](https://www.licklider.ai/engineering/games-howell-approximation-and-guarantees/): source-reviewed GH, T2/T2-prime, T3/C distinctions; simulations are not universal error guarantees.
- [Control or best](https://www.licklider.ai/engineering/comparing-with-control-or-best/): source-reviewed targets, ordered testing and interval boundaries.
- [Testing graphs](https://www.licklider.ai/engineering/implementing-multiple-testing-graphs/): source-reviewed closure, ordering, weights and endpoint conventions. Release 3 method adoption and public support remain separate decisions from the experiments.

- [Multiple-testing source review](https://www.licklider.ai/engineering/checking-multiple-testing-against-original-papers/): six original papers reviewed; Rom table/equation conflict confirmed in the project review. Historical Release 3 source evidence; the bounded supplied-source RFC is now open for discussion. Bounded SR-C acceptance is recorded; later SR-F, SR-I, SR-D and SR-J source results also have scoped acceptance. Historical counts describe their pinned snapshots. Overall sources remain incomplete; no Release 3 support is announced.

- [Research](https://www.licklider.ai/research/): papers and research notes
- [Engineering](https://www.licklider.ai/engineering/): implementation work, upstream reports, and distinct repair outcomes
- [Blog](https://www.licklider.ai/blog/): practical guidance for researchers using AI
- [Latest](https://www.licklider.ai/latest/): all public updates in reverse chronological order
- [RSS](https://www.licklider.ai/rss.xml): RSS 2.0 feed for all public updates
- [JSON Feed](https://www.licklider.ai/feed.json): JSON Feed 1.1 for all public updates