Agenda Intel MD

v0.9.0 Feature

This release adds 3 notable features for engineering teams evaluating rollout.

Published 14d MCP Developer Tools

View tool

✓ No known CVEs patched

Read the diff → Tool health → What is this tool? →

✓ No known CVEs patched in this version

Topics

a2a a2a-protocol agenda-intelligence agent-infrastructure agentic-workflow ai-agents

+14 more

cloudflare-workers deal-risk-gate decision-grade geopolitical-intelligence geopolitical-risk json-schema mcp mcp-server model-context-protocol policy-analysis regulatory-compliance risk-analysis sanctions strategic-intelligence

Summary

AI summary

Updates What changed in v0.9.0, Other Unreleased work merged, and Product-shell narrative alignment across a mixed release.

Changes in this release

Type	Severity	Summary	CVE
Feature
Feature	Medium	Agent-Eval Delta introduced as structural product validation surface in v0.9. Agent-Eval Delta introduced as structural product validation surface in v0.9. Source: granite4.1:8b-q6_K@2026-05-20 Confidence: high	—
Feature	Medium	ADOPTION section lists 16 product-shell tools alongside validation tools. ADOPTION section lists 16 product-shell tools alongside validation tools. Source: granite4.1:8b-q6_K@2026-05-20 Confidence: high	—
Feature	Medium	Request-context prompt guardrails added to analyze tool. Request-context prompt guardrails added to analyze tool. Source: granite4.1:8b-q6_K@2026-05-20 Confidence: high	—
Feature	Medium	Product-shell contract alignment with 16-tool MCP surface across README and tables. Product-shell contract alignment with 16-tool MCP surface across README and tables. Source: granite4.1:8b-q6_K@2026-05-20 Confidence: high	—
Feature	Medium	New glossary entries added for Agent-Eval Delta and Practitioner Review in CONTEXT.md. New glossary entries added for Agent-Eval Delta and Practitioner Review in CONTEXT.md. Source: granite4.1:8b-q6_K@2026-05-20 Confidence: low	—
Refactor
Refactor	Medium	README headline reworded to describe MCP product shell and evidence-discipline layer. README headline reworded to describe MCP product shell and evidence-discipline layer. Source: granite4.1:8b-q6_K@2026-05-20 Confidence: low	—
Refactor	Medium	Source Acquisition Layer renamed to Source Planning Layer in llms.txt and packaged copy. Source Acquisition Layer renamed to Source Planning Layer in llms.txt and packaged copy. Source: granite4.1:8b-q6_K@2026-05-20 Confidence: low	—
Refactor	Medium	agent-eval case template and index files created in evals/agent-eval directory. agent-eval case template and index files created in evals/agent-eval directory. Source: granite4.1:8b-q6_K@2026-05-20 Confidence: low	—

Full changelog

What changed in v0.9.0

Agent-eval delta as v0.9 validation surface

ROADMAP v0.9 rewritten from verify-quotes network mode improvements to agent-eval delta and product-shell narrative alignment. Verify-quotes patches deferred to a new v0.9.x section.
New glossary entries in CONTEXT.md: Agent-Eval Delta (per-case structural delta from the agent-integrator perspective) and Practitioner Review (optional, audience-gated).
docs/adr/0008-agent-eval-delta-is-structural-product-validation.md records the validation-story decision.
docs/agent-eval-methodology.md tightened: live-source-backed examples map to user_provided or mixed for analyze; live_source_backed is intentionally absent from agenda-request.schema.json because live retrieval is upstream of Agenda Intelligence.
evals/agent-eval/ scaffolds three cases (gtta-global-policy.md intended as the full case; ca-caspian-sanctions.md and gulf-me-hormuz-shipping.md as stubs) plus reusable TEMPLATE.md.

Product-shell narrative alignment

README headline drops "trusted geopolitical intelligence layer" framing for "MCP product shell and evidence-discipline layer for strategic intelligence agents". Explicit non-goals: no live retrieval, no factual verification before v1.0.
ADOPTION lists product-shell tools (analyze, validate_memo, list_signals, get_signal, deep_dive) alongside the 11 validation tools.
llms.txt and packaged copy: Source Acquisition Layer renamed to Source Planning Layer to match CONTEXT canon. The source_acquisition manifest key is unchanged (compatibility wire name).

Other Unreleased work merged

Request-context prompt guardrails in analyze (server-verified context block; no invented citations when no source material is present).
Product-shell contract alignment: 16-tool MCP surface in both README overview and table; analyze honors output_format: markdown; sanctions-sector uses sector_specialist role.
Agent-eval case template (TEMPLATE.md) and evals/agent-eval/README.md index.

Install

pip install agenda-intelligence-md==0.9.0

Full changelog: CHANGELOG.md

View diff on GitHub

Weekly OSS security release digest.

The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.

No spam, unsubscribe anytime.

Share this release

Share on X Share on Bluesky

Track Agenda Intel MD

Get notified when new releases ship.

About Agenda Intel MD

All releases →

Related context

Related tools

Earlier breaking changes

v0.8.0 MCP tool count increased from 11 to 16, adding five new tools.