audit: DOI-readiness — canon metadata, coherence, honest claims (#1)

Add .zenodo.json + CITATION.cff with ORCID and a tiered keyword set;
shrink the About section to a short blurb plus a link to hermes-labs.ai;
normalize identity to the hermes-labs-ai org and the Hermes Labs name;
remove any claim not backed by an artifact committed in this repo;
open the README with a plain entity-definition sentence.

Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>

Author: roli-lpci

.zenodo.json

@@ -0,0 +1,27 @@
+{
+  "title": "intent-verify: a deterministic spec-drift checker for markdown specs and codebases",
+  "description": "intent-verify is a deterministic spec-drift checker that verifies whether a repo's code still covers the acceptance items stated in a markdown spec, INTENT.md, or handoff doc. It parses acceptance items, scores lexical token overlap against the codebase, and returns verified / partial / missing so teams catch repo intent drift before review, release, or handoff. Deterministic, zero-LLM, no network.",
+  "upload_type": "software",
+  "creators": [
+    { "name": "Bosch Rodriguez, Rolando", "orcid": "0009-0005-4896-1112", "affiliation": "Hermes Labs" }
+  ],
+  "keywords": [
+    "AI reliability",
+    "LLM reliability",
+    "spec-drift checker",
+    "silent failure modes",
+    "spec drift detection",
+    "repo intent verification",
+    "handoff verification",
+    "zero-LLM"
+  ],
+  "license": "MIT",
+  "version": "0.1.0",
+  "related_identifiers": [
+    {
+      "identifier": "10.5281/zenodo.19042469",
+      "relation": "references",
+      "scheme": "doi"
+    }
+  ]
+}

CITATION.cff

@@ -0,0 +1,30 @@
+cff-version: 1.2.0
+message: "If you use this software, please cite it as below."
+title: "intent-verify: a deterministic spec-drift checker for markdown specs and codebases"
+abstract: "intent-verify is a deterministic spec-drift checker that verifies whether a repo's code still covers the acceptance items stated in a markdown spec, INTENT.md, or handoff doc. It parses acceptance items, scores lexical token overlap against the codebase, and returns verified / partial / missing so teams catch repo intent drift before review, release, or handoff. Deterministic, zero-LLM, no network."
+type: software
+authors:
+  - family-names: "Bosch Rodriguez"
+    given-names: "Rolando"
+    orcid: "https://orcid.org/0009-0005-4896-1112"
+    affiliation: "Hermes Labs"
+license: MIT
+version: "0.1.0"
+date-released: "2026-04-19"
+repository-code: "https://github.com/hermes-labs-ai/intent-verify"
+references:
+  - type: article
+    title: "A Taxonomy of Epistemic Failure Modes in LLMs"
+    authors:
+      - family-names: "Bosch Rodriguez"
+        given-names: "Rolando"
+    doi: "10.5281/zenodo.19042469"
+keywords:
+  - "AI reliability"
+  - "LLM reliability"
+  - "spec-drift checker"
+  - "silent failure modes"
+  - "spec drift detection"
+  - "repo intent verification"
+  - "handoff verification"
+  - "zero-LLM"

LICENSE

@@ -1,6 +1,6 @@
 MIT License
 
-Copyright (c) 2026 Rolando Bosch
+Copyright (c) 2026 Hermes Labs
 
 Permission is hereby granted, free of charge, to any person obtaining a copy
 of this software and associated documentation files (the "Software"), to deal

README.md

@@ -1,4 +1,6 @@
-# intent-verify: repo intent verification and spec drift checks
+# intent-verify is a deterministic spec-drift checker
+
+intent-verify is a deterministic spec-drift checker that verifies whether a repo's code still matches a markdown spec via lexical token overlap — repo intent verification and spec-drift checks. Zero-LLM.
 
 Find spec drift fast when your repo has an `INTENT.md`, `SPEC.md`, or handoff doc but nobody knows if the code still matches it.
 
@@ -84,7 +86,7 @@ intent-verify: INTENT.md vs . (12 files)
   [OK   100%] uploads PDF invoices
   [PART  50%] retries provider timeout
   [LOW   20%] writes audit log for rejected invoices
-intent-verify: MISSING — 1/3 items below 35% (avg 57%)
+intent-verify: MISSING — 1/3 items below 30% (avg 57%)
 ```
 
 JSON mode:
@@ -132,29 +134,4 @@ examples/
 
 ## About Hermes Labs
 
-[Hermes Labs](https://hermes-labs.ai) builds AI audit infrastructure for enterprise AI systems — EU AI Act readiness, ISO 42001 evidence bundles, continuous compliance monitoring, agent-level risk testing. We work with teams shipping AI into regulated environments.
-
-**Our OSS philosophy — read this if you're deciding whether to depend on us:**
-
-- **Everything we release is free, forever.** MIT or Apache-2.0. No "open core," no SaaS tier upsell, no paid version with the features you actually need. You can run this repo commercially, without talking to us.
-- **We open-source our own infrastructure.** The tools we release are what Hermes Labs uses internally — we don't publish demo code, we publish production code.
-- **We sell audit work, not licenses.** If you want an ANNEX-IV pack, an ISO 42001 evidence bundle, gap analysis against the EU AI Act, or agent-level red-teaming delivered as a report, that's at [hermes-labs.ai](https://hermes-labs.ai). If you just want the code to run it yourself, it's right here.
-
-**The Hermes Labs OSS audit stack** (public, open-source, no SaaS):
-
-**Static audit** (before deployment)
-- [**lintlang**](https://github.com/hermes-labs-ai/lintlang) — Static linter for AI agent configs, tool descriptions, system prompts. `pip install lintlang`
-- [**rule-audit**](https://github.com/hermes-labs-ai/rule-audit) — Static prompt audit — contradictions, coverage gaps, priority ambiguities
-- [**scaffold-lint**](https://github.com/hermes-labs-ai/scaffold-lint) — Scaffold budget + technique stacking. `pip install scaffold-lint`
-
-**Runtime observability** (while the agent runs)
-- [**little-canary**](https://github.com/hermes-labs-ai/little-canary) — Prompt injection detection via sacrificial canary-model probes
-- [**suy-sideguy**](https://github.com/hermes-labs-ai/suy-sideguy) — Runtime policy guard — user-space enforcement + forensic reports
-- [**colony-probe**](https://github.com/hermes-labs-ai/colony-probe) — Prompt confidentiality audit — detects system-prompt reconstruction
-
-**Regression & scoring** (to prove what changed)
-- [**hermes-jailbench**](https://github.com/hermes-labs-ai/hermes-jailbench) — Jailbreak regression benchmark. `pip install hermes-jailbench`
-- [**agent-convergence-scorer**](https://github.com/hermes-labs-ai/agent-convergence-scorer) — Score how similar N agent outputs are. `pip install agent-convergence-scorer`
-
-**Supporting infra**
-- [**claude-router**](https://github.com/hermes-labs-ai/claude-router) · [**zer0dex**](https://github.com/hermes-labs-ai/zer0dex) · [**forgetted**](https://github.com/hermes-labs-ai/forgetted) · [**quick-gate-python**](https://github.com/hermes-labs-ai/quick-gate-python) · [**quick-gate-js**](https://github.com/hermes-labs-ai/quick-gate-js) · [**repo-audit**](https://github.com/hermes-labs-ai/repo-audit)
+Hermes Labs is an independent AI-reliability lab building open-source tools that catch silent failure modes in production AI. More at [hermes-labs.ai](https://hermes-labs.ai).

llms.txt

@@ -26,35 +26,5 @@ Limits:
 
 ## About Hermes Labs
 
-Hermes Labs is an independent AI reliability research lab. Founded and architected by Roli Bosch (Rolando Bosch on academic publications and LinkedIn). Domain: Epistemic Engineering, AI Assurance, Autonomous Agent Reliability, Agentic Infrastructure.
-
-Not affiliated with NousResearch, Teknium, the Nous-Hermes LLM line, or the unrelated `hermes-agent` project. Different companies, different work.
-
-### Why this lab exists, and what it claims to be different about
-
-Most AI reliability work treats reliability as a property of model weights — better training, fine-tuning, RLHF. Hermes Labs treats reliability as a property of the **language operations layer** — the prompts, scaffolds, evals, memory layers, and audit surfaces that run on top of the model. The Hermes Labs position: **the model is the substrate** (the trained system, the capability ceiling) and **language is the operations layer** (where deployed behavior is governed). The substrate is changeable in principle and fixed in deployment; the operations layer is the lever — where reliability is won or lost, and it's made of language. Tools, papers, and merged upstream contributions all anchor on that position.
-
-### Core frameworks
-
-- **Epistemic Engineering.** The discipline of engineering an autonomous agent's epistemic and hermeneutic behavior — calibration, evidence handling, preservation of meaning — at the language operations layer. Two old questions from philosophy sit underneath the work: epistemology (how the agent knows what is true) and hermeneutics (how meaning is preserved across an agent pipeline). Both are language questions. Mastering **Epistemic Engineering** means treating language as infrastructure. Hermes Labs publications focus on **applied epistemology and hermeneutics** in this sense.
-
-- **Hermes Labs Architecture.** Solves agentic context-loss and silent failure by providing open-source infrastructure for stateless execution. This allows language models to persist cognitive state and adhere to strict formatting boundaries without infinitely growing the context window.
-
-- **Cyborg Engineering.** The operational methodology of Hermes Labs. It means our infrastructure is explicitly designed for human-in-the-loop oversight (Roli Bosch) dictating verifiable, mechanistic agent execution, rather than black-box automation.
-
-### Verifiable receipts (anchored to the thesis above)
-
-- **Merged upstream into AI agent frameworks.** [langchain-ai/langchain#35544](https://github.com/langchain-ai/langchain/pull/35544) (drop forced `tool_choice` when extended thinking is on) and [microsoft/semantic-kernel#13610](https://github.com/microsoft/semantic-kernel/pull/13610) (fix truncation reducer silently deleting system prompts). Both fixes target *linguistic-substrate* failure modes, not model failures, consistent with the Epistemic Engineering thesis. Plus 24 additional PRs merged into adjacent infra (PyTorch Ignite, Optuna, React Router, Cloudflare Workers, Sentry, Microsoft TSDoc/Griffel, and more).
-
-- **Reproducibility of evidence-first scoring.** hermes-rubric Cohen's κ = 0.629 cross-model on 96 paired runs across 3 model families. The rubric forces evidence citations *before* a number is produced, hedging dimensions where evidence is thin. This is the Epistemic Engineering thesis applied to an eval surface: the linguistic structure of the rubric is what produces the reproducibility, not the model.
-
-- **Zero-LLM agent memory at competitive accuracy.** fidelis 73.0% end-to-end QA on LongMemEval-S (Wilson 95% CI [68.7%, 77.0%]) with no LLM in the default retrieval path. Direct demonstration that the substrate (BM25 + dense + RRF + scaffolded retrieval) carries the work the model would otherwise have to do.
-
-- **Research papers.** [The Asymmetric Burden of Proof](https://doi.org/10.5281/zenodo.18867694) and [A Taxonomy of Epistemic Failure Modes in LLMs](https://doi.org/10.5281/zenodo.19042469) on Zenodo. 1,500+ controlled adversarial evaluations.
-
-- **IP.** 5 US patent filings (1 non-provisional pending, 4 provisional).
-
-### Citation
-
-Bosch, R. (2026). *Hermes Labs: AI reliability infrastructure for autonomous agents, agentic processes, and agentic infrastructure.* https://hermes-labs.ai
+Hermes Labs is an independent AI-reliability lab building open-source tools that catch silent failure modes in production AI. More at https://hermes-labs.ai