Owliver could offer neither create. The Create Position flow worked and no chip
anywhere suggested it, because the chip row is entirely the backend's static
catalogue and no intent in it wrote anything. The gap was never in the
frontend's trigger matching — every phrasing already routed.
`employee_roles` is the supply side of `job_postings`. A posting is what the
ORGANIZATION needs filled; this is what a WORKER says they do. They share a
vocabulary and almost nothing else: "3 years" on a posting is a minimum an
applicant must clear, and the same words here are what the person has. There is
deliberately no foreign key between them — supply and demand already meet
through `job_applications`, which carries the funnel, the interview and the
outcome, and a second weaker link would disagree with it the first time
somebody withdrew.
NO NEW COMPANY ENTITY, AND THAT IS THE LOAD-BEARING DECISION. "Create a company
position" reads like it needs a client record. `organizations` is the TENANT —
absent from the resource table, absent from the policy map, written only by the
seeder — so creating a row there from a chat flow would provision a new tenant,
and the position would carry an org_id the operator's session cannot see. The
operator could never view the record they just created. That breaks I5 and I1
to add a feature nobody asked for. The client stays free text on the posting,
per blueprint decision D2, and the flow simply offers the clients this
organization already staffs for as chips. No schema change, no endpoint change.
Create is operators-only, and that is an I1 decision rather than a deferral.
The worker is named explicitly on the row and is deliberately NOT derived from
the session, because an operator recording a role on somebody's behalf is the
whole point of the flow. Granting talent the same Create would let a talent
caller write a role under any worker_email in the tenant — the attribution hole
Phase 3D closed elsewhere. Talent reads its own via a ScopeEmail predicate,
which is in place now so the grant is one line when a talent console exists.
`created_by` is in gen_resources.py's SERVER_OWNED as well as the policy's
Derived list. Both are required and the pairing is easy to miss: Derived fills
the column from the session, SERVER_OWNED is what makes the descriptor ReadOnly
so a request body cannot set it in the first place. Without it,
TestDerivedColumnsAreReadOnlyOrTalentScoped fails — verified by mutation, not
by reading.
The two catalogue intents carry PHRASE terms only. A bare "position" or "role"
term scores 10, the same as every reading on that page, and wins the tie on
declaration order — so a create chip would have arrived by evicting
`positions-attention` from the exact ordered result TestPositionsSuggestions
asserts. An offer to create something must not displace the reading a person
actually asked for. Neither declares a Subject, on the precedent of
`position-spec-steps`: a Subject would let the bare query "summarize" match
through matchShape and survive filterOnTopic. Neither declares a Signal, so an
empty composer still reports what the organization needs rather than proposing
paperwork.
Chip text is the coupling with nothing else holding it together: no page
context declares `capabilities`, so every server suggestion dispatches as its
own TEXT and is answered by whichever skill's trigger that text matches. A
renamed chip would open nothing, silently. Asserted on the frontend side.
The down migration drops `employee_role_status` and keeps `english_level`,
which is shared with job_postings.english_required and
job_applications.english_level. Rolled back and re-applied against the
database to prove it, not asserted.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PJvibeSc1JYXjatankqM1g
Correcting the previous commit's reasoning. It claimed the difference between
this file and what production published was inert — a defaulted `webSearch:
false` and a reordered skills list. That was based on comparing the file
against the LIVE row in agent_definitions. The live row was the wrong thing to
compare against: Snapshot compares against the stored v1 in
definition_versions, and those two had diverged.
Read from production directly, v1 as published carries two skills:
skills:
- anomaly-detection
- operational-risk
and the live row carries three, with activity-analysis added. So a skill was
added to this agent after v1 was published, in place, without the version being
raised — the exact silent rewrite this branch exists to stop. It is a genuine
change of behaviour: an agent with a third skill answers differently from one
with two.
So the version is raised rather than the file being bent to match. v1 keeps
what was approved; the current definition, which is what production has been
serving, becomes v2. The earlier alignment of field order and `webSearch:
false` is kept — that part WAS serialisation, and matching it keeps future
imports quiet.
Production has three agents with version rows at all, so two others may hold
the same kind of drift. They did not conflict on this import, which means their
live rows still match what was published; it does not mean nobody edited them.
The image ships agents/, so this file only reaches production on the next
image build. Any build from main at or after this commit carries it; a build
from an older tree will fail the import again, by design.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PJvibeSc1JYXjatankqM1g
The first run of the new importagents against production refused, which is the
guard working rather than a fault: version 1 of activity-agent was already
published and said something different from the file this repository ships.
The difference is inert. Production's copy adds `webSearch: false`, which is
exactly what the parser defaults to when the key is absent (agent.go:471 reads
`data["webSearch"] == true`), and lists the same three skills in a different
order. Both forms parse to the identical agent — 2 tools, 0 sources, 3 skills,
confirmed by running the importer's own dry-run over each.
What happened is a round trip: somebody edited this agent in the UI, the editor
re-serialised it, and that serialisation is what got snapshotted as v1. The
hand-authored file was never the published artefact for this one.
So the file is updated to match rather than the version being bumped. Bumping
would publish a v2 that differs from v1 only in field order and a defaulted
key, which is noise in a history whose whole purpose is to say what changed.
This unblocks the import. It does not address the underlying awkwardness: the
comparison is textual, so any future UI edit that reformats without changing
meaning will block a deploy the same way. Comparing the PARSED definition
instead would fix that properly and is the right follow-up — it needs a
decision about what counts as semantically equal (skill order, starter order)
and should not be rushed in behind a deploy.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PJvibeSc1JYXjatankqM1g
