view release on metacpan or  search on metacpan

lib/Chorus/Engine/AIAgent.pod  view on Meta::CPAN

regulations…) for each pipeline agent.

=item * B<Six AI agent skills> — reusable prompt-driven workflows
(C<chorus-pdf>, C<chorus-feed>, C<chorus-create-project>,
C<chorus-import-project>, C<chorus-check>, C<chorus-strengthen>) that drive
an AI agent to build and run complete validation applications from a corpus.

=back

The resulting workflow is entirely deterministic: the rules are derived from
the corpus, encoded as transparent YAML files, and executed verbatim by the
Chorus inference engine — no LLM in the hot path.

=head1 OVERVIEW

The architecture connects three layers:

  ┌──────────────────────────────────────────────────────────┐
  │  Normative corpus  (PDF, DTU, Eurocode, regulation…)     │
  └────────────────────────┬─────────────────────────────────┘
                           │  chorus-pdf (optional, PDF input)

lib/Chorus/Expert.pm  view on Meta::CPAN

(C<_SUCCES> is true on that agent).  This implements priority-based sequencing:
earlier agents are given another full pass before the locked agent is allowed
to run.

Typical use: a "global cleanup" or "conformity check" agent that should only
run once all upstream agents have stabilised for the current cycle.

=head2 _REPLAY and _REPLAY_ALL

These flags are set by the corresponding engine methods and are handled
transparently by C<process()>.

=over 4

=item C<_REPLAY>

Set by C<< $agent->replay() >>.  C<process()> re-runs C<loop()> on the same
agent immediately (inner C<do/while> loop), without advancing to the next agent.

=item C<_REPLAY_ALL>