feat: working TOM agent
This commit is contained in:
10
.agents/plans/README.md
Normal file
10
.agents/plans/README.md
Normal file
@@ -0,0 +1,10 @@
|
||||
# Plans
|
||||
|
||||
One directory per iteration. `../../IMPLEMENTATION_PLAN.md` is the plan across this repository and the engine's; its steps T1 onward are built here.
|
||||
|
||||
| # | Directory | Status | What it is |
|
||||
| --- | --- | --- | --- |
|
||||
| 1 | `iter-01-first-slice/` | the Workflow, its checks and views built on 2026-10-13 from a written operating spec; the digest and evaluation remain | Six Deliverables, from an operating spec to a priced, validated operating model |
|
||||
| 2 | `iter-02-platform-deployment/` | steps 1 to 3 built 2026-10-20; step 4 in progress | The agent as a container on the local KAB platform, then behind the Orchestration engine |
|
||||
|
||||
Each iteration has a `MAP.md` (intent, each step's status, sources of truth) and `steps/step-N.md`, written when a step is finished. A step with no file is not done.
|
||||
31
.agents/plans/iter-01-first-slice/MAP.md
Normal file
31
.agents/plans/iter-01-first-slice/MAP.md
Normal file
@@ -0,0 +1,31 @@
|
||||
# Plan Map
|
||||
|
||||
**Iteration:** 1, the first slice
|
||||
**Intent:** From an operating spec to a priced, validated operating model, in six Deliverables.
|
||||
**Status:** T1 to T14 built; both evaluation cases passed with real models on 2026-10-14
|
||||
|
||||
## Sources of truth
|
||||
|
||||
- `.scratch/target-operating-model/spec.md`: the formats, the rules and the arithmetic.
|
||||
- `docs/adr/`: 0001 (the graph, and findings), 0002 (the operating spec), 0003 (cost).
|
||||
- `IMPLEMENTATION_PLAN.md`, section 10: the steps.
|
||||
- `../orchestration_engine/execution_engine/workflows/workflow.py`: what a Workflow supplies.
|
||||
|
||||
## Status
|
||||
|
||||
| Step | What | Status |
|
||||
| --- | --- | --- |
|
||||
| T1 | The repository on the engine | built (`steps/step-1.md`) |
|
||||
| T2 | The operating catalogue and scoping Skill | built (`steps/step-1.md`) |
|
||||
| T3 | The architecture adapter and the digest command | built (`steps/step-2.md`) |
|
||||
| T4 | The model-drafted digest for other documents | built as instructions to the scope analyst (`steps/step-2.md`); not yet run by a model |
|
||||
| T5 | The capability catalogue Skill and its sources register | built (`steps/step-1.md`); references unverified |
|
||||
| T6 | Operational requirements and capabilities, and their checks | built (`steps/step-1.md`) |
|
||||
| T7 | The graph: refusals and rendered views | built (`steps/step-1.md`) |
|
||||
| T8 | Findings | built (`steps/step-1.md`) |
|
||||
| T9 | Role files, the provider bundle, the Workflow | built (`steps/step-1.md`) |
|
||||
| T10 | Validation in code: verdict rules and scenario walk-throughs | built (`steps/step-1.md`) |
|
||||
| T11 | The guide | built (`steps/step-2.md`) |
|
||||
| T12 | The rate card Skill's format and checks | built (`steps/step-1.md`) |
|
||||
| T13 | Staffing and cost in code | built (`steps/step-1.md`) |
|
||||
| T14 | Evaluation with real models | built; two cases passed (`steps/step-2.md`) |
|
||||
44
.agents/plans/iter-01-first-slice/steps/step-1.md
Normal file
44
.agents/plans/iter-01-first-slice/steps/step-1.md
Normal file
@@ -0,0 +1,44 @@
|
||||
# Steps T1, T2, T5 to T13: the Workflow, from a written operating spec
|
||||
|
||||
Built 2026-10-13, on the Execution engine as it stood that day (not yet committed there).
|
||||
|
||||
## What was built
|
||||
|
||||
- **The Workflow** (`app/operating_model/workflow.py`): five Phases, six Deliverables, each with its Specialist, sections and data file; the validation policy; the views; the account of a Change request; the card.
|
||||
- **The operating spec's catalogue** (`skills/operating-scoping/`): 32 fields, read by the engine's spec reader. The provider bundle supplies the Cloud provider and the currencies.
|
||||
- **The capability catalogue** (`skills/operating-capabilities/`): 17 entries in five domains, each citing frameworks from a register of nine. `catalogue.py` refuses an entry that cites nothing or cites an unregistered framework, and proposes the entries that apply to a spec.
|
||||
- **The chain** (`chain.py`): the checks on operational requirements and on capabilities.
|
||||
- **The graph** (`graph.py`): its shape for the engine's graph tools, its refusals, findings, and the rendered responsibility matrix, organisation chart, and staffing and cost.
|
||||
- **Staffing and cost** (`staffing.py`): the rate card's loader and checks, and the arithmetic.
|
||||
- **Scenarios** (`scenarios.py`): five kinds, generated and traced by code, their verdicts written into `verdicts.json`.
|
||||
- **The agent's folder**: eight role files and a shared protocol, `providers/gcp.yaml`, the `operating-chain` Skill, `app/main.py`.
|
||||
- **106 tests**, among them seven whole Executions on the scripted runtime.
|
||||
|
||||
## Differed from the plan and the spec
|
||||
|
||||
- **The spec's files keep the engine's names,** `scope.yaml` and `scope.json`. The plan said `operating-spec.yaml`. The engine's spec support does not yet let a Workflow name them.
|
||||
- **The inventory travels with the rest of the spec** to every Specialist, not as a separate `solution.json`. The engine's digest step (E6) is where that changes.
|
||||
- **The graph holds no list of Capabilities.** They are the capability model's, named by id, so a failed Check can reopen who does the work without reopening what the work is.
|
||||
- **The currencies a spec may state are listed in the provider bundle,** so the engine refuses another at the start. A test keeps the list the same as the rate card's conversion rates.
|
||||
- **A Role on business hours is as small as its work.** The spec's formula applied the minimum roster to every Role; the plan, which the owner approved, applies it only to hours wider than a working day. The plan was followed.
|
||||
- **A Check that depends on a finding is recorded as `waits_on_finding`.** The validator gives it `not_verifiable`; code rewrites the verdict when the validation is submitted. Left as `not_verifiable`, the engine's router reopened the capability model for it, again and again: a finding is not a gap in the model.
|
||||
- **A scenario's verdicts are written into `verdicts.json` by code,** as Checks whose ids begin `SC-`, and the walk-through as a whole is the file's `dry_run`. A failed scenario is therefore a failed Check to the router, which reopens the graph.
|
||||
- **The matrix shows both parts** when one Role is accountable and responsible for a Capability (`A/R`).
|
||||
|
||||
## Not built
|
||||
|
||||
- **The digest** (T3, T4). An Execution starts from an operating spec that is already written. Nothing reads the solution architecture agent's files yet.
|
||||
- **A guide template** (T11). The packager has its sections and its instructions.
|
||||
- **Evaluation** (T14). Nothing has run with a real model: no role file, Skill or catalogue entry has been read by one.
|
||||
- An interactive Execution and a Change request are not tested here; the engine's own tests cover both on other Workflows.
|
||||
- The capability catalogue's references are drafted from general knowledge and marked `unverified`. Nobody has checked them against the frameworks.
|
||||
- The rate card's figures are placeholders.
|
||||
|
||||
## What it found in the engine
|
||||
|
||||
- The engine was not typed as a package (`py.typed` added there), so this repository could not be type-checked against it.
|
||||
- An editable install of the engine is invisible to the type checker unless made with `editable_mode=compat`.
|
||||
|
||||
## Verified
|
||||
|
||||
`pytest -q`: 106 passed. `ruff check .`, `ruff format --check .`, `mypy .` (20 files): clean. The engine's suite still passes (521). `git status` in the solution architecture agent shows the same nine modified files as before any of this work began.
|
||||
38
.agents/plans/iter-01-first-slice/steps/step-2.md
Normal file
38
.agents/plans/iter-01-first-slice/steps/step-2.md
Normal file
@@ -0,0 +1,38 @@
|
||||
# Steps T3, T4, T11 and T14: the digest, the guide, and the first runs with real models
|
||||
|
||||
Built 2026-10-13 and 14, with the engine's digest step (its `steps/step-6.md`).
|
||||
|
||||
## What was built
|
||||
|
||||
- **The architecture adapter** (`app/operating_model/digest.py`, `python -m app.operating_model.digest <folder>`). An Item is a component of the architecture, with the Elements that realise it; a Condition is what its requirements and recommendations say can only be shown in operation (`outside_code`); a Check that did not pass is an inherited risk. Only what the customer stated in the architecture's scope is carried across. The provider bundle's `operating.item_kinds` says what kind of thing each product is. Tested against a real completed landing-zone Execution, copied into `tests/fixtures/architecture/`.
|
||||
- **Documents code cannot read** go to the scope analyst, who records what it takes from them with the passage quoted (T4, as instructions; no model has run it).
|
||||
- **The guide** (`app/operating_model/guide.py`, `skills/operating-guide/`). The packager writes the summary; code writes every other section and the key figures from the data.
|
||||
- **The evaluation harness** (`eval/`): two cases, structural scoring, tokens per role.
|
||||
|
||||
## Differed from the plan
|
||||
|
||||
- An Item is a component, not an Element. The real architecture had 60 Elements in 9 components; an organisation operates the components.
|
||||
- The architecture's 104 Checks on infrastructure settings are not Conditions. Keeping them true is configuration management. Its 12 statements of what can only be shown in operation are.
|
||||
|
||||
## The runs with real models, 2026-10-14
|
||||
|
||||
| Case | Result | Minutes | Revisions | Capabilities | Roles | Checks | Scenarios | People | Cost a month |
|
||||
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
|
||||
| 01, written spec (4 Items) | pass | 25.8 | operating model ×2, capability model ×1 | 17 | 4 | 8 | 13 of 13 | 5.2 | 25,355 GBP |
|
||||
| 02, digested architecture (9 Items) | pass | 16.8 | operating model ×1 | 17 | 5 | 36 | 31 of 31 | 11.6 | 67,298 GBP |
|
||||
|
||||
Costs are from the placeholder rate card. An earlier run of case 01 passed every structural check in 30.6 minutes, over its limit then of 30.
|
||||
|
||||
What they showed:
|
||||
|
||||
- **Most time goes on the architect placing a Check on the wrong Capability**, such as a restore-time Check on service-level reporting. Case 01 revised the same way in both runs. Not yet acted on.
|
||||
- **The router's reasons still speak of code.** A Check that could not be verified reopens the capability model with "could not be verified from the code", and a Check that failed twice is said to be possibly "not implementable". These are the engine's words from the architecture's subject.
|
||||
- **The capability modeller takes all 17 catalogue entries** whatever the size of the solution, as its instructions say.
|
||||
- **The architect is most of the cost:** about 380,000 to 490,000 of each case's 780,000 to 940,000 prompt tokens.
|
||||
- One request was retried after Vertex AI's tokens-per-minute quota was reached; it did not affect the result.
|
||||
|
||||
Nobody has yet read the outputs for quality: these checks are structural only.
|
||||
|
||||
## Verified
|
||||
|
||||
The engine's 529 tests and this repository's 140 pass, with `ruff` and `mypy` clean in both.
|
||||
67
.agents/plans/iter-01-first-slice/steps/step-3.md
Normal file
67
.agents/plans/iter-01-first-slice/steps/step-3.md
Normal file
@@ -0,0 +1,67 @@
|
||||
# Socio-technical and organisational frameworks
|
||||
|
||||
Added 2026-10-14, at the owner's request.
|
||||
|
||||
## What was added
|
||||
|
||||
- **Six frameworks in the register** (`skills/operating-capabilities/assets/sources.yaml`): Team Topologies; platform as a product, cited from the CNCF Platforms White Paper; the SPACE framework; and the DevEx framework. The DORA delivery metrics were already there. Like every other entry, their references are unverified.
|
||||
- **Four Capabilities in the catalogue:** platform product management, self-service and golden paths, and enablement, proposed for a platform foundation; and delivery and developer experience measurement, proposed for a platform foundation or a workload. The catalogue can now propose an entry by the kind of engagement.
|
||||
- **Measurement at three layers.** The measurement Capability cites DORA, SPACE and DevEx together, each for a different layer: delivery speed and stability as the outcome, sustainable productivity as the guardrail, friction as the leading signal. The Skill says no one layer stands in for another.
|
||||
- **Teams and interactions in the graph,** both optional. A Team is of one of Team Topologies' four kinds, in one Party; a Role may be in a Team of its own Party. An Interaction between two Teams is in one of the three modes, and a collaboration must say when it ends. The organisation chart draws teams inside their Parties and interactions as labelled edges; the guide lists both.
|
||||
- **Two new terms** in `CONTEXT.md`: Team and Interaction.
|
||||
- **The architect's instructions** say how to use teams and interaction modes, and that a Check goes on the Capability that does the work, not the one that reports it. The second was the cause of most Revisions in the runs of 2026-10-14.
|
||||
|
||||
## Not done
|
||||
|
||||
- No run with real models since. The four new Capabilities add to what the capability modeller takes, which was already all 17 entries for a platform foundation; it will now be 21.
|
||||
- Teams are not used by staffing, cost or the scenario walk-throughs.
|
||||
|
||||
## Verified
|
||||
|
||||
159 tests pass; `ruff` and `mypy` clean.
|
||||
|
||||
## The run with real models after this change, 2026-10-14
|
||||
|
||||
| Case | Minutes | Revisions | Capabilities | Teams | Interactions | Checks | Scenarios | People | Cost a month | Model cost |
|
||||
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
|
||||
| 01, written spec | 20.9 (was 25.8) | operating model ×1 (was 3 in all) | 21 | 2 | 2 | 13 | 13 of 13 | 6.6 | 32,234 GBP | $1.87 (was $2.27) |
|
||||
| 02, digested architecture | 16.6 (was 16.8) | operating model ×1 (was 1) | 21 | 3 | 2 | 31 | 35 of 35 | 12.5 | 77,550 GBP | $1.61 (was $1.55) |
|
||||
|
||||
Both passed every check. The one Revision in each was still a Check whose hours or Capability did not fit, but the capability model was not reopened in either. Both models used teams: platform and stream-aligned teams, an X-as-a-Service interaction in each, and in case 01 a collaboration that ends three months after go-live, in case 02 a facilitating interaction that ends with the build. Case 02 named six Parties.
|
||||
|
||||
# Business outcomes
|
||||
|
||||
Added 2026-10-14, at the owner's request: every measure is tied back to an outcome the business stated.
|
||||
|
||||
- **`intent.business_outcomes`** in the operating spec: each with an id, the customer's words, and a target where given. The digest carries the architecture's stated success criteria across as business outcomes.
|
||||
- **Measures** in the graph: what is measured, its layer (outcome, guardrail, leading, service level or cost), its framework, its target, the outcome it shows and the Capability that takes it. Refused: a stated outcome shown by no Measure, a Measure for an outcome that was not stated, a Measure taken by a Capability nobody is responsible for, a framework not in the register, and, where the model measures how work flows, any of the three developer layers left unmeasured.
|
||||
- **An Operational requirement may come from a business outcome.** It need not: the Measures are what show each outcome is served.
|
||||
- **The guide's second section is Business outcomes,** written by code: each outcome with the Measures that show it and the Role answerable for each.
|
||||
- **Two new terms** in `CONTEXT.md`: Business outcome and Measure. The benchmark spec now states three outcomes.
|
||||
|
||||
Not yet run with real models. 173 tests pass; `ruff` and `mypy` clean.
|
||||
|
||||
## The run with business outcomes, 2026-10-14
|
||||
|
||||
- **Case 01 passed** in 29.2 minutes, for $2.73. It tied four Measures to the spec's three business outcomes: deployment lead time (DORA, outcome layer) to faster releases, time to restore service (DORA) to short outages, and satisfaction and wellbeing (SPACE, guardrail) and feedback loops and cognitive load (DevEx, leading) both to keeping engineers. Two Revisions of the operating model and one of the capability model.
|
||||
- **Case 02 looped and was stopped** after about two and a half hours. The architecture states no business outcome, so every Measure was refused as serving an outcome "none were stated", while the measurement Capability's rule asked for Measures at three layers. The two rules could not both be met. The architect asked a question five times in a pre-approved run, each answered automatically, each run taking 10 to 28 minutes, and the last model request hung. About 173 Pro requests; an estimated $7 to $10.
|
||||
|
||||
Fixed after:
|
||||
|
||||
- `intent.business_outcomes` is essential. With someone present it is asked for; a pre-approved start must state them or leave them to the agent with `decide`. Case 02 now does the latter.
|
||||
- The three-layer rule applies only when there is an outcome to tie Measures to.
|
||||
- The evaluation harness stops a case at one and a half times its time limit and records it as stopped.
|
||||
|
||||
Not fixed: the engine lets a pre-approved Specialist end without submitting again and again, each time answered automatically, until the Orchestrator's turn limit. A guard there would have stopped case 02 within minutes.
|
||||
|
||||
## Case 02 rerun, and findings kept to the solution, 2026-10-14
|
||||
|
||||
Case 02 passed with its business outcomes left to the agent: 31.6 minutes, $2.71, two Revisions. The agent assumed three outcomes from the request (the landing zone ready within 10 weeks, continuous DORA and PCI DSS compliance, high availability with low-latency hybrid connectivity) and tied six Measures to them, all three developer layers on the first.
|
||||
|
||||
It also filed a gap in its own model as a finding: service continuity staffed only in business hours against a 4-hour recovery target. That drafted a change request to the architecture agent for a fault that was the operating model's. Fixed:
|
||||
|
||||
- **A finding names the Items of the solution it is about.** One with none is refused.
|
||||
- **The validator judges each finding.** A Check waiting on a finding the operating model could meet itself (staffing, hours, a Role, a process, a tool) is given `fail`, which sends the model back to the architect. Only a Check left `not_verifiable` is recorded as waiting on its finding.
|
||||
- **The architect is told** what a finding is and is not.
|
||||
|
||||
177 tests pass. Not yet run with real models.
|
||||
31
.agents/plans/iter-02-platform-deployment/MAP.md
Normal file
31
.agents/plans/iter-02-platform-deployment/MAP.md
Normal file
@@ -0,0 +1,31 @@
|
||||
# Plan Map
|
||||
|
||||
**Iteration:** 2, on the platform
|
||||
**Intent:** The operating model agent runs as a container beside the local KAB platform, is registered in its catalogue, and completes an Execution started by a platform workflow. The Orchestration engine (F1, F2, in `../orchestration_engine`) is then put in front of it.
|
||||
**Status:** steps 1 to 3 built 2026-10-20; step 4 in progress
|
||||
|
||||
## Decided by the owner, 2026-10-20
|
||||
|
||||
- **E7, locally:** the image is built with a named build context, `--build-context engine=../orchestration_engine`, and installs the Execution engine from it. No file is copied by hand. CI will use a dependency on the engine's repository at a pinned commit.
|
||||
- **Order:** this agent first (steps 1 to 4), then the Orchestration engine (step 5).
|
||||
- **Models:** Gemini on Vertex AI, as in the evaluation, with the developer's application default credentials mounted read-only. The platform's LiteLLM is not used.
|
||||
|
||||
## Sources of truth
|
||||
|
||||
- `IMPLEMENTATION_PLAN.md`, sections 2 and 10 (E7, F1, F2).
|
||||
- `../modernization-platform/packages/agent-builder-local-dev-kit/README.md`: the local platform, `agents-compose.yml` and `scripts/compose.sh agents`.
|
||||
- `../modernization-platform/packages/agent-builder-temporal-workers/src/temporal/a2a_client.py`: how a platform workflow calls an agent.
|
||||
|
||||
## Known mismatch
|
||||
|
||||
A platform workflow step treats an A2A `input_required` as the agent runtime's action review (a list of tool calls to approve, at most `MAX_ACTION_REVIEWS` rounds). The engine's questions and Approvals are not that, so a platform workflow calls this agent pre-approved. Interactive Executions go through the Orchestration engine, which relays every prompt (F2). Step 4 confirms this rather than assuming it.
|
||||
|
||||
## Status
|
||||
|
||||
| Step | What | Done when | Status |
|
||||
| --- | --- | --- | --- |
|
||||
| 1 | The image (E7): `Dockerfile`, `.dockerignore`, the engine from the named build context | The image builds from a clean checkout of both repositories; `/health` and the agent card answer | built (`steps/step-1.md`) |
|
||||
| 2 | A real-model run through the container | One pre-approved Execution of case 01 completes through the container's A2A endpoint | built (`steps/step-2.md`) |
|
||||
| 3 | On the platform's network: an entry in the dev kit's `agents-compose.personal.yml`, registered in the catalogue | It is listed in the catalogue; the API container reaches `http://tom-analysis:8080/health` | built (`steps/step-3.md`) |
|
||||
| 4 | A platform workflow with one step calling this agent | The workflow completes in Temporal with the Deliverables in its output | in progress: the engine's card declares KAB's input contract (`request`, `customer_case`, `scope`, `pre_approved`), chosen by the owner; waiting on the catalogue refresh and the workflow |
|
||||
| 5 | The Orchestration engine in front (F1, F2 in `../orchestration_engine`) | An Execution of either agent runs through it, Approvals relayed | F2 built (`../../../orchestration_engine/.agents/plans/iter-02-orchestration-engine/steps/step-2.md`): this agent's questions relayed through it on the platform; F1 awaits the owner |
|
||||
29
.agents/plans/iter-02-platform-deployment/steps/step-1.md
Normal file
29
.agents/plans/iter-02-platform-deployment/steps/step-1.md
Normal file
@@ -0,0 +1,29 @@
|
||||
# Step 1: the image (E7)
|
||||
|
||||
Built 2026-10-20.
|
||||
|
||||
## What was built
|
||||
|
||||
- **`Dockerfile`.** Two stages. The builder installs the Execution engine from a named build context, `engine`, copying only its `pyproject.toml` and `execution_engine/`. The runtime stage holds `app/`, `agents/`, `providers/` and `skills/` in `/app` and runs `python -m app.main` as a non-root user (10001) on port 8080. The agent is not installed: `ROOT` in `workflow.py` is found from `__file__`, so it must run from its own folder, as the reference agent does.
|
||||
- **Defaults for a container:** `CHECKPOINTER=sqlite` with its files in the `/data` volume, and `ENV_FILE_EXPORT=false`.
|
||||
- **`.dockerignore`:** keeps `.venv`, `.cache`, `.env`, `eval/results`, `deliverables`, tests and plans out of the image.
|
||||
|
||||
```bash
|
||||
podman build --build-context engine=../orchestration_engine -t localhost/agents/tom-analysis:local .
|
||||
```
|
||||
|
||||
## What differed from the plan
|
||||
|
||||
- **Credentials and the non-root user.** The application default credentials file is mode 600, so the image's user cannot read it under rootless podman. The container is run with `--userns=keep-id:uid=10001,gid=10001`, which maps the developer's user onto the image's. The file's mode is not changed.
|
||||
- **Podman ignores `HEALTHCHECK`** in an OCI-format build. The compose entry declares its own (step 3).
|
||||
|
||||
## Verified
|
||||
|
||||
- Neither repository has a commit yet, so "a clean checkout" was taken as the files git would commit (`git ls-files --cached --others --exclude-standard`) of both, copied to an empty folder. The image built from that, and the Workflow loads in it with its five Phases.
|
||||
- The container answers `/health` and both agent card paths; the card advertises `A2A_PUBLIC_URL`. From inside it, the credentials are readable and a Gemini request on Vertex AI answers.
|
||||
- An A2A `message/send` with `pre_approved`, `customer_case` and `scope` in the message metadata reached the engine, which checked the spec. A Customer case that differs from the spec's is refused by name.
|
||||
- 177 tests pass; `ruff` and `mypy` are clean. No Python changed.
|
||||
|
||||
## Not done
|
||||
|
||||
- CI's means, a dependency on the engine's repository at a pinned commit, needs the engine's repository to have a commit to pin.
|
||||
20
.agents/plans/iter-02-platform-deployment/steps/step-2.md
Normal file
20
.agents/plans/iter-02-platform-deployment/steps/step-2.md
Normal file
@@ -0,0 +1,20 @@
|
||||
# Step 2: a real-model run through the container
|
||||
|
||||
Run 2026-10-20.
|
||||
|
||||
## What was run
|
||||
|
||||
Evaluation case 01's written spec (`eval/datasets/specs/landing-zone-shared.yaml`), sent to the container from step 1 as one A2A `message/send` with `configuration.blocking: false`, then polled with `tasks/get`. The message metadata carried `pre_approved: true`, `customer_case: bench-landing-zone-shared` and the spec as `scope`. Models were Gemini on Vertex AI with the developer's application default credentials, mounted read-only.
|
||||
|
||||
## Result
|
||||
|
||||
| Minutes | State | Capabilities | People | Scenarios |
|
||||
| --- | --- | --- | --- | --- |
|
||||
| 31.6 | completed | 20 | 5.1 | 13 of 13 pass |
|
||||
|
||||
The task's artifacts were all eleven files: the guide, `scope.yaml` and `scope.json`, operational requirements, capabilities, the Target operating model and its organisation chart, staffing, findings, verdicts and scenarios.
|
||||
|
||||
## What differed from the plan
|
||||
|
||||
- **The Customer case must be the spec's.** The first attempt named another slug and was refused at once: "The scope spec is for bench-landing-zone-shared, and this was started for smoke-01." A caller that sends a spec sends its Customer case with it.
|
||||
- **Postgres.** The engine tries `localhost:5432`, logs a warning and falls back to SQLite in `/data`. Harmless here; a shared deployment sets `DATABASE_URL`.
|
||||
28
.agents/plans/iter-02-platform-deployment/steps/step-3.md
Normal file
28
.agents/plans/iter-02-platform-deployment/steps/step-3.md
Normal file
@@ -0,0 +1,28 @@
|
||||
# Step 3: on the platform's network
|
||||
|
||||
Built 2026-10-20.
|
||||
|
||||
## What was built
|
||||
|
||||
In the dev kit (`../modernization-platform/packages/agent-builder-local-dev-kit`), two personal files, excluded from git in that clone's `.git/info/exclude`:
|
||||
|
||||
- **`agents-compose.build.personal.yml`:** `tom-analysis` built from `../../../tom_analysis` with `additional_contexts: {engine: ../../../orchestration_engine}`.
|
||||
- **`agents-compose.personal.yml`:** the container `tom-analysis` on `agent-builder-network`, port `${TOM_ANALYSIS_PORT:-10020}:8080`, `A2A_PUBLIC_URL=http://tom-analysis:8080`, the credentials mounted read-only, named volumes for `/data` and `/app/deliverables`, and its own healthcheck.
|
||||
|
||||
```bash
|
||||
./scripts/compose.sh agents build tom-analysis
|
||||
./scripts/compose.sh agents up tom-analysis
|
||||
```
|
||||
|
||||
The owner registered it in the catalogue from the admin page (Agents, Discover agent, `http://tom-analysis:8080`). It is listed as `target_operating_model_agent`.
|
||||
|
||||
## What differed from the plan
|
||||
|
||||
- **No pod.** podman-compose puts every container in one pod, and podman refuses a per-container user namespace (`keep-id`, step 1) inside a pod. The personal file sets `x-podman: in_pod: false`. Agents reach each other by name on the network, not through the pod.
|
||||
- **The platform calls the address an agent was registered with,** not the A2A interface its card names (`agent_catalog_service.py` sets `endpoint.url` to the discovery URL). The engine served JSON-RPC only at `/a2a`, so every platform call would have been a 404. Chosen by the owner: the engine serves the same handler at `/` as well (`execution_engine/main.py`, with a test in `tests/toy/test_server.py`). The card still names `/a2a`. The platform's choice is worth raising with its team.
|
||||
|
||||
## Verified
|
||||
|
||||
- From the API and workers containers: `/health`, both card paths, and JSON-RPC at `/` and at `/a2a`.
|
||||
- The catalogue record names the agent, its skill and its examples.
|
||||
- The engine: 532 tests pass; `ruff` and `mypy` clean. This agent: 177 tests pass.
|
||||
Reference in New Issue
Block a user