Resources: the 45 cards
The 45 ACF® toolkit cards exposed as MCP resources, one per card, at acf://fiche/ACF-XX: Decide (ACF-01 to ACF-20), Organise (ACF-O1 to ACF-O10), Operate (ACF-P1 to ACF-P15). Each card is served as Markdown, in French and English.
acf.advisor tool proposes a sequence of cards from a free-text description.What this contains
45 resources, one per card, of type text/markdown. Each card carries the title, subtitle, purpose and sections of the printable card published on acfstandard.com, with the download link. The acf.fiche.lookup tool returns the same content with the PDF link and, on request, the metadata.
Exhaustive list ×45
acf://fiche/ACF-01MarkdownFR + ENACF-01 · Sovereignty Score
Measure your exposure before automating
acf://fiche/ACF-02MarkdownFR + ENACF-02 · Decision Map
Map your decisions and their level of autonomy
acf://fiche/ACF-03MarkdownFR + ENACF-03 · Criticality Matrix
Which decisions can be delegated to an agent?
acf://fiche/ACF-04MarkdownFR + ENACF-04 · Agentic Constitution
The rules, limits and responsibilities of the agent
acf://fiche/ACF-05MarkdownFR + ENACF-05 · Agent Card
Frame every deployed agent
acf://fiche/ACF-06MarkdownFR + ENACF-06 · Supervision & Governance
Monitor, measure and keep the agent compliant
acf://fiche/ACF-07MarkdownFR + ENACF-07 · Kill Switch
Emergency shutdown protocol
acf://fiche/ACF-08MarkdownFR + ENACF-08 · First Agent Dossier
Document, frame and prepare the deployment of your first agent
acf://fiche/ACF-09MarkdownFR + ENACF-09 · Agentic Decisions Register
Log significant decisions and incidents to ensure traceability and compliance
acf://fiche/ACF-10MarkdownFR + ENACF-10 · Action Plan & Improvement
Steer corrective actions and continuous improvement of your agents
acf://fiche/ACF-11MarkdownFR + ENACF-11 · 30-Day Governance Audit
Assess the maturity of your agentic governance and build a 30-day action plan
acf://fiche/ACF-12MarkdownFR + ENACF-12 · Agentic Risk Assessment
Identify, assess and prioritize risks tied to your AI agents to control them more effectively
acf://fiche/ACF-13MarkdownFR + ENACF-13 · Agent Mandate
Define the scope, limits and responsibilities of every deployed agent
acf://fiche/ACF-14MarkdownFR + ENACF-14 · Responsibility by Design
Responsibility by Design · embedding accountability from inception
acf://fiche/ACF-15MarkdownFR + ENACF-15 · Objective and Counter-Metric
Judging the target against an out-of-reach figure
acf://fiche/ACF-16MarkdownFR + ENACF-16 · Supervision Test
Testing the arrangement, never the person
acf://fiche/ACF-17MarkdownFR + ENACF-17 · Validity Review
Judging whether a mandate still holds
acf://fiche/ACF-18MarkdownFR + ENACF-18 · Guided Case Study
Apply the ACF® framework to a real-world scenario, step by step
acf://fiche/ACF-19MarkdownFR + ENACF-19 · Instructor's Guide
Facilitate, grade and evaluate agentic governance workshops
acf://fiche/ACF-20MarkdownFR + ENACF-20 · Governance Simulation
Train to decide, arbitrate and supervise in real-world conditions
acf://fiche/ACF-O1MarkdownFR + ENACF-O1 · Human · AI · Agent Matrix
Who acts, who is accountable, which rule, who stays in the loop
acf://fiche/ACF-O2MarkdownFR + ENACF-O2 · Hybrid Organisation Map
Placing humans, agents and rules, domain by domain
acf://fiche/ACF-O3MarkdownFR + ENACF-O3 · Critical Skills to Keep
Keeping the ability to judge your agents
acf://fiche/ACF-O4MarkdownFR + ENACF-O4 · AI Policy
The text that says what is done with AI, before the first agent
acf://fiche/ACF-O5MarkdownFR + ENACF-O5 · Shadow AI, Undeclared Usage
The tools nobody approved and everybody uses
acf://fiche/ACF-O6MarkdownFR + ENACF-O6 · Shadow Agents, Undeclared Agents
The ones the organisation never deployed, already deciding in its name
acf://fiche/ACF-O7MarkdownFR + ENACF-O7 · The Shift of Power
Who lost a decision, who gained it, with nobody deciding it
acf://fiche/ACF-O8MarkdownFR + ENACF-O8 · The Exceptions Register
What was allowed once, and has lasted ever since
acf://fiche/ACF-O9MarkdownFR + ENACF-O9 · Governing the Fleet
Governing twenty agents is not governing one agent twenty times
acf://fiche/ACF-O10MarkdownFR + ENACF-O10 · Affected People and Their Appeal
Whoever bears the decision must be able to challenge it before someone
acf://fiche/ACF-P1MarkdownFR + ENACF-P1 · Agent Configuration Grid
Setting the agent's six dials
acf://fiche/ACF-P2MarkdownFR + ENACF-P2 · Acceptance Protocol
Proving every dial before the first customer
acf://fiche/ACF-P3MarkdownFR + ENACF-P3 · Go-Live Ramp
Pilot → steps → rollback
acf://fiche/ACF-P4MarkdownFR + ENACF-P4 · Weekly Supervision Log
The four rhythms that keep hold of the agent
acf://fiche/ACF-P5MarkdownFR + ENACF-P5 · Role Allocation
Who does what, on the ground
acf://fiche/ACF-P6MarkdownFR + ENACF-P6 · Escalation & Confidence Card
Knowing when, and how, to hand over
acf://fiche/ACF-P7MarkdownFR + ENACF-P7 · Stop Mechanism Test
The kill switch proven every quarter
acf://fiche/ACF-P8MarkdownFR + ENACF-P8 · Incident Log
Detect → root cause → fix
acf://fiche/ACF-P9MarkdownFR + ENACF-P9 · Permissions & Data Review
Least privilege, reviewed
acf://fiche/ACF-P10MarkdownFR + ENACF-P10 · Vendor Assessment Grid
The five questions before buying an agent
acf://fiche/ACF-P11MarkdownFR + ENACF-P11 · Foundation Dependency Map
What stops if access stops
acf://fiche/ACF-P12MarkdownFR + ENACF-P12 · Degraded Mode and Exit Plan
Deciding without them, then leaving
acf://fiche/ACF-P13MarkdownFR + ENACF-P13 · Model Version Log
What reopens the mandate
acf://fiche/ACF-P14MarkdownFR + ENACF-P14 · Holding the Record
Evidence you can still produce
acf://fiche/ACF-P15MarkdownFR + ENACF-P15 · Blind Spot Statement
What the framework does not establish
Fetching a card
A resources/read call on the card URI. The language is chosen with the ?lang=fr or ?lang=en parameter, otherwise with the ACF_MCP_LOCALE environment variable, English by default. The response starts with a header stating the language actually served.
// Read ACF-13, Agent Mandate, in French
const { contents } = await client.readResource({
uri: "acf://fiche/ACF-13?lang=fr",
});
console.log(contents[0].mimeType); // "text/markdown"
console.log(contents[0].text);
// ---
// served_locale: fr
// is_fallback: false
// ---
//
// # Mandat d'Agent
//
// *Définir le périmètre, les limites et les responsabilités de chaque agent déployé*
// ...