Changelog

Unreleased

• Correct current UK source routing and preserve unverified topic status.
• Align evidence modes, calibration, and regional reference loading.
• Restore substantive documentation checks and harden specialist evaluation.

All notable changes to this repository are documented here.

Unreleased

• Removed the mock streaming engine and all claims of autonomous enforcement, continuous graph traversal, persistent memory, or no-human-review operation.

• Replaced the enforcement-shaped JSON schema with a non-enforcing reviewer recommendation contract that always requires human review.

• Added a transport-independent MCP fallback contract and a standard-library CI test proving unavailable tools return not_implemented and never clearance.

• Corrected MCP docstrings and documented the graph and memory files as proposed data shapes rather than implemented services.

• Moved the complete runtime-neutral analytical contract into the root SKILL.md and reduced the Claude and Codex files to additive overlays.

• Replaced the packaged Claude symlink with a composition adapter whose name matches its directory. The validator now enforces the root/overlay split, package identity, attachment order, and plugin-manifest parity.

• Added a structural Claude/Codex runtime-loading smoke record. It covers instruction loading and bounded safety behavior only; it is not a factuality or model-quality benchmark.

• Removed unsupported production-stack and retainer wording from the public README. OpenClaw remains deliberately deferred.

• Fixed a dead link to the author's site: the contact section listed for-analysts.html, which has returned 404 since the site was restructured. It now links the live case-study page for this skill.

• Corrected the repository name casing in evals/README.md (Agenda-Intelligence-md -> agenda-intelligence-md); it resolved only through a GitHub redirect.

• Extended scripts/check_markdown_links.py to fail on 404/410 for links to the author's own site, which previously went unchecked because every http(s):// target was skipped. Network errors and other statuses are reported without failing; SKIP_SITE_LINK_CHECK=1 skips the network step.

• Conformed the repo to the Agent Plugins 1.0.0 layout: added a root plugin.json with the $schema identifier from https://agent-plugins.org. .claude-plugin/plugin.json is unchanged and still serves the Claude Code install path; the specification ignores that directory. skills/gulf-middle-east/SKILL.md already matched the spec's discovery rule. The manifest validates against the published Draft 2020-12 schema.

• Added an Agenda Intelligence v1.3 evidence-packet handoff, a runnable synthetic regional packet, a dependency-free CI validator, and ADR 0001. Reclassified the older analyze / memo-validation composition as compatibility behavior.

• Renamed the runtime-overlay directory skills/{claude,codex}/ to runtimes/{claude,codex}/ and updated every path reference (README, AGENTS, CLAUDE, CONTRIBUTING, STATUS, evals, examples, validator, CI workflow). skills/ is now reserved for Claude Code plugin packaging, because plugin installs auto-discover every skills/*/SKILL.md as a separate skill and the overlay layout produced junk-named skills (claude, codex).

• Packaged the repo as an installable Claude Code plugin: added .claude-plugin/plugin.json and skills/gulf-middle-east/SKILL.md (a symlink to the canonical root SKILL.md). Installable via /plugin marketplace add vassiliylakhonin/agenda-intelligence-md, then /plugin install gulf-middle-east@agenda-intelligence.

• Clarified the repo's commercial role: Gulf + Middle East remains a regional specialist reasoning layer, while Gulf maritime / trade-finance readiness is only a backup Agenda Intelligence MD hypothesis. Added README/AGENTS guidance to avoid new buyer-facing surfaces or deployed workers without buyer evidence.

• Added the unified portfolio stack-map table to README.md section 7, with the vertical-specialist row bolded for this repo.

• Added a "Project maturity" callout under section 7 pointing readers to STATUS.md (Bar 1 / Bar 2 framework) and AGENTS.md "Definition of done".

• Added CONTEXT.md (regional vocabulary: Iran-actor distinctions, chokepoint scope, sovereign wealth deployment terms, evidence-mode boundary versus the product shell).

• Added .github/ISSUE_TEMPLATE/{bug_report,feature_request}.md and .github/pull_request_template.md so contributors have a structured entry point.

• Added this CHANGELOG.

• Cross-referenced the portfolio-wide skill packaging convention in AGENTS.md "Relationship to the broader stack".