Two bugs from review, the second in both agents because identity.py is a copy.
The exchanged-token cache grew for the life of the process. It is keyed on
(subject token, scope) and the subject token is minted fresh every turn, so
every turn added keys nothing would ever read again — and expiry was checked on
read but never collected. A slow leak that only shows in a pod up for days,
which is the worst way to find one.
It now drops expired entries first and evicts oldest beyond a bound. Expired
first on purpose: an expired entry is dead weight wherever it sits in the
ordering, so evicting by age alone would keep it while discarding a live one.
The docstring was wrong before the code was: it claimed the cache lasted one
turn. It never did. Corrected rather than quietly made true.
message_id reused the context id, which is the conversation — and a conversation
makes several calls to the same peer, relaying an approval being exactly that.
A peer treating the second call as a replay of the first would surface as a
stale answer rather than an error.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The agent asked for confirmation because the system prompt told it to, and read
the answer itself. Both halves are now enforced.
Each MCP toolset is wrapped in ApprovalRequiredToolset whose predicate reads the
readOnlyHint the server publishes — the gateway decides what counts as a write,
and the agent keeps no list of tool names in step with it.
output_type=[str, DeferredToolRequests] lets a turn end holding a proposed call
instead of making it; the executor parks the A2A task in input-required and
renders the arguments the model actually sent, not what it said it would do.
The reply is matched against a closed set of words, so "oui mais change le prix"
denies the call and the correction comes back for approval on its own merits —
verified end to end, along with a draft the user approved that was still refused
because msrp sat below buyprice.
Also wired: a ProductDraft validated after approval and before the call leaves;
UsageLimits on every run; FilteredToolset denying sales' write tools inside the
agent as well as at the gateway; PrefixedToolset namespacing the two servers;
dynamic instructions naming the caller; Agent.instrument_all() for a span per
model call and per tool call; and pydantic_ai.retries on inference only, so a
429 from the gateway is waited out rather than failing mid-reasoning.
Inference gets its own httpx client: validate_response raises on any 4xx, and
applying that to the shared client would turn a token-exchange refusal into an
exception carrying no body — which is where Keycloak says what was wrong. Its
transport is built explicitly, because a transport passed to the client makes
verify= inert and the lab CA would otherwise go undeclared.
Tests grew from 7 to 31, covering what approval decides and what a draft must
satisfy. evals/catalogue_evals.py adds a behavioural suite, kept out of pytest
because every case spends real tokens; its first case checks that a count comes
from the tool rather than from what the model remembers of a public dataset.
CLAUDE.md corrected: instrument=True is not an Agent parameter in 2.35.3,
pydantic_ai.retries needs tenacity, TemplateStr needs pydantic-handlebars, and
approval does not remove the in-memory history — resuming still needs the
messages, so replicas: 1 stands until they live in the A2A task store.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
First of the six agents, and the one that gives the A2A facade for the
others to interoperate with. A turn exchanges the user's token once per
audience, builds one MCP toolset per server with those tokens, runs the
model through the Kong AI gateway, and drops the toolsets.
Per-turn MCP clients are the point, not an implementation detail: a token
that outlives its turn can be replayed after the session ended. Several
narrow tokens rather than one broad one, so a prompt injection that talks
the agent into the wrong domain is refused at Kong rather than succeeding
quietly.
a2a-sdk rather than FastA2A, revising the May decision. That decision
rested on Agent.to_a2a() being removed in PydanticAI v2 — which has now
happened, so the argument no longer separates them. With four other
runtimes due to interoperate, the protocol author's implementation is the
safer base. httpx2 rather than httpx, which PydanticAI deprecates for
OpenAI-compatible providers and drops in v3.
Four things about a2a-sdk 1.1 that no documentation gave and that were
established against the running library, all recorded where they bite:
a2a.types are protobuf, not pydantic; DefaultRequestHandler speaks 0.3 and
V2 is needed for 1.0; a client must send A2A-Version: 1.0 or the server
assumes 0.3 and answers VERSION_NOT_SUPPORTED on a perfectly valid
request; and an executor must enqueue a Task before any status update.
Verified: the app builds with no deprecation warnings, serves its Agent
Card, and fails a turn cleanly with "no bearer token" rather than
answering from the model's own knowledge — which would hide a broken
identity chain behind a plausible reply. Seven tests cover the exchange
shape (scope not audience, per-audience caching, refusal) and the card.
The Keycloak client does not exist yet, so no end-to-end run has happened.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Stéphane Tailland