docs(news): state what a reader gets, and move the design where it is read

The v1.36.0 release note carried four things a reader cannot use. The dashboard's
seven in-browser tools and the installtype fix stay; the withheld judgement verbs,
the consequentialHint limitation and the two silent-failure notes are removed from
the release feed and moved to where an implementer looks.

  - The four verbs are withheld because an agent that can approve can approve its
    own work. That is settled design, so it now sits in docs/architecture/identity.md,
    beside the approval-path conflict it belongs with, stated as the constraint it is:
    attribution, not capability, is what releasing them depends on.
  - The browser-tool surface had no home in docs/architecture, so it has one now:
    docs/architecture/dashboard-agent-tools.md holds what a tool call can be
    attributed to, why confirmation is a signal rather than a control, and the two
    ways the surface fails silently.

Audited 1.35.0, 1.36.0 and 1.37.0 for the same class. 1.37.0 is clean. 1.35.0 carried
two build notes: a bullet recounting that a guard had never been written because it
failed, and a clause about a manager "finally" having a production writer. Both are
gone; the capabilities they were attached to remain.

RELEASING.md now carries the rule so the next release reads it: release notes state
what a reader gets, and design constraints, withheld capabilities and failure modes
belong in docs/architecture/, because a note about an absence reads as a feature to
someone who never had the thing.

docs/news/** is regenerated by task news, never hand-edited. task check passed before
this commit: the chain runs it, so a failing gate commits nothing.
This commit is contained in:
sam committed 2026-09-22 16:53:27 +10:00
1 parent 674c9816a0
commit c43bd46ecb
8 files changed
+77 -62

No files matched your search

+1 -29
View File
@@ -57,29 +57,6 @@
agent's call and a page script's call reach the server identically. Tools therefore record the
signed-in identity and a source, and **never** that an agent acted.
### The four judgement verbs are deliberately absent
- `approve`, `reject`, `abandon` and `archive` are **not** exposed. `internal/domain/taskactions`
records only a source string and stamps `KindHumanApproved`, so an agent's approval would be written
down as a person's. An agent that can approve can approve its own work, and the one control this
design exists to keep is a human accepting it. Those verbs land once task actions carry both the
acting identity and the authorising principal.
### `consequentialHint` does not gate anything, and the release says so
- Chrome 152 accepts `consequentialHint` at registration and **omits it from `getTools()`**, so no
client sees it and the browser cannot enforce what it does not surface. The three recovery tools
instead require an explicit `confirm: true`, which is a **deliberate-action signal rather than a
security control** an agent can trivially satisfy. The boundary that holds is the session: the tools
cannot exceed what the signed-in operator can already do.
### Two ways this fails silently, now recorded next to the code
- **A secure context is required.** On a plain-HTTP origin `document.modelContext` is absent *even
with the flag*, and nothing reports why. `mkcert` covers local development.
- **`archie-ui` embeds `ui/dist` into its binary**, so rebuilding the distribution is not deploying it;
a running process can serve a bundle with no tools while the built one has them.
### Binaries built by the development build task can be updated
- `task build` passed the `installtype` stamp for three of six commands, so `archie-messaging`,
@@ -97,10 +74,6 @@
`internal/app/controlplane`; the client half now lives in
`internal/infrastructure/controlplanerpc`, and the app package keeps the server, the registry, the
seeds, the validators and the schema derivation.
- **The PRD's guard now exists, and passes:** `cmd/archie-messaging/architecture_test.go` measures the
linked binary against the banned list, with `internal/domain/workflow/task` as an exception checked
in both directions. Before this the guard had never been written -- because it failed. A messaging
process linking a SQLite driver and the workflow engine is a severance hole, not a design choice.
- The resource documents are type-aliased back into the app package, so there is one definition read
under two names rather than a copy per side.
@@ -108,8 +81,7 @@
- The Messaging Service reports each channel's state, recorded **per channel**: one channel's failure
marks only itself, a channel that cannot start is marked failed **with its reason** rather than only
logged, and one that returns is marked stopped. `internal/channels/status.Manager` finally has a
production writer.
logged, and one that returns is marked stopped. `internal/channels/status.Manager` reports each channel's state.
- `PutChannelStatus`/`ListChannelStatus` carry that to the dashboard, following the config-snapshot
split. The store **replaces the reported set**, so a channel the reporter stops mentioning is
deleted rather than left as a stale `running`; an empty report clears the table, because hosting no
+9 -1
View File
@@ -41,7 +41,15 @@ needs the new agent image to actually take effect, say so explicitly and
either cut an agent release or note that `docker compose build agent` /
`docker compose pull agent` is needed separately.
## Process
## What the notes are for
**Release notes state what a reader gets.** A reader scanning releases does not
need to know what was withheld, what a browser API cannot do, or what the
engineering took; and a note about an absence reads as a feature to someone who
never had the thing. Design constraints, withheld capabilities and failure modes
belong in `docs/architecture/`, where implementers look, because there they are
requirements rather than news.
```bash
task release:preview VERSION=1.3.0 # preview what would land, both components
@@ -0,0 +1,39 @@
# The dashboard's tools for an in-browser agent
The dashboard registers in-browser tools an agent can call through the page's model
context. Four are read-only — `list_tasks`, `get_task`, `recent_events`,
`daemon_health` — and three recover stuck work: `retry_task`, `stop_task`,
`cancel_task`. Each wraps the same `ui/src/lib/api.ts` client the dashboard itself
calls, so an agent and a person exercise one path rather than two.
The four judgement verbs are withheld for the reason `identity.md` gives: an agent
that can approve can approve its own work.
## What a tool call can be attributed to
A page can invoke its own tools, so an agent's call and a page script's call reach
the server identically — the difference does not exist on the wire, and no header
recovers it. A tool call is therefore attributed to the signed-in identity the
request presented, with its source recorded as a hint; it never records that an
agent acted, and it never records that a human approved. An agent's action is
attributable only on a surface where the agent presents its own credential.
## Confirmation is a signal, not a control
The three recovery tools require an explicit `confirm: true`, which makes a
consequential call deliberate rather than incidental. It is not a security control:
an agent can trivially satisfy it, and the browser cannot enforce
`consequentialHint` at all — Chrome accepts it at registration and omits it from
`getTools()`, so no client can see what the browser does not surface.
The boundary that holds is the session. These tools cannot exceed what the operator
signed in at the provider is already permitted to do.
## Two ways the surface fails silently
- **A secure context is required.** On a plain-HTTP origin `document.modelContext` is
absent even with the flag enabled, and nothing reports why. `mkcert` covers local
development.
- **The dashboard embeds `ui/dist` into its binary.** Rebuilding the distribution is
not deploying it: a running process can serve a bundle without tools while the
built one has them, or serve a bundle older than its own binary.
+14
View File
@@ -263,6 +263,20 @@ identity.
## Current enforcement and isolation
## Withheld judgement verbs
`approve`, `reject`, `abandon` and `archive` are not exposed to an agent. An agent
that can approve can approve its own work, and the one control this surface exists
to keep is a human accepting it.
The constraint is attribution, not capability: a task action records the acting
identity and the authorising principal, so exposing the verbs requires that an
agent's action be recorded as an agent's and the authority it used recorded
separately. Until a surface carries both, the verbs stay withheld. This is the
reason the four verbs are absent from the dashboard's in-browser tools, and it is
why releasing them is a change to this boundary rather than a change to the tools.
The current implementation uses identity equality as an authorization check for
chat task approval and cancellation. This conflicts with the approved rule that
identity is attribution rather than authentication or access control and must be
+11 -1
View File
@@ -30,6 +30,12 @@
"section": "Architecture",
"body": "# Configuration\n\n**Status:** Approved; existing-type assignment is ongoing \n**Date:** 2026-07-28 \n**Tracking issue:** [#73](https://github.com/samcharles93/archie-core/issues/73)\n\n## Ownership\n\nConfiguration has two separate responsibilities:\n\n1. `internal/infrastructure/configuration` loads files, environment variables,\n secret references, overlays, and compatibility representations.\n2. Each domain and plugin owns the typed runtime settings required for its\n behaviour.\n\n`internal/app` translates external input into those settings, validates\ncross-capability combinations, and supplies each component only its own\nsettings.\n\nThe following are prohibited:\n\n- passing complete application configuration into a domain;\n- making a domain depend on TOML, YAML, environment-variable names, secret\n references, or a configuration library;\n- defining a central runtime configuration type imported across the system;\n- placing plugin options in an unrelated global configuration model.\n\n## Private input document\n\n`internal/infrastructure/configuration` MAY use a complete decoded document\nwhile discovering files, decoding sections, and applying overlays. It is an\ninput-boundary DTO only.\n\nThe document:\n\n- remains private to the configuration infrastructure;\n- is composed from separately defined feature sections;\n- is immediately translated by application composition;\n- is never returned as the application's runtime model;\n- is never passed through runtime components.\n\nExternal DTOs and domain runtime settings are separate types.\n\nPlugins supply external configuration definitions through a narrow extension\ncontract. The core loader MUST NOT gain a field or recognized filename for every\nplugin.\n\nInvalid configuration input MUST NOT replace active valid settings or terminate\nunaffected capabilities. Configuration changes use the candidate validation,\npromotion, audit, and recovery requirements in\n[safe-change-and-recovery.md](/dev/architecture/safe-change-and-recovery/).\n\n## Dissolution of `internal/config`\n\n`internal/config` has no target location. It MUST be dissolved and MUST NOT\nsurvive as a shared package, configuration domain, or renamed global model.\n\nIts loading behaviour moves to `internal/infrastructure/configuration`. Existing\ntypes and methods are reassigned by behaviour. Repository execution policy,\ndispatch policy, retry policy, ecosystem defaults, tool policy, memory settings,\nindexing settings, provider settings, and channel settings move to their\nrespective owners.\n\n## Repository settings\n\nThe existing `config.Repo` MUST NOT move intact into a repository domain or\nshared settings package. A managed repository is an application association:\n\n- work intake owns which repository references an identity monitors;\n- task execution owns gates, preflight, test protection, ecosystem behaviour,\n and change limits;\n- task lifecycle owns retry behaviour;\n- scheduling owns concurrency policy;\n- workspace management owns base branch and persistent-storage requirements.\n\nApplication composition correlates these settings by repository reference. No\ndomain receives a universal repository profile.\n\n## Documentation\n\nGenerated configuration reference describes external sections and their owning\ncapabilities. It MUST NOT establish a global Go configuration structure as the\npublic contract.\n\n## Value classification rule\n\nEvery value that controls behaviour MUST have an explicit owner and MUST be\nclassified as exactly one of:\n\n1. **Invariant** — changing it changes the meaning of the program, a wire\n protocol, or a domain concept. It may be a Go constant.\n2. **Runtime setting** — an operator, identity, domain, plugin, or deployment\n may legitimately choose a different value. It is typed configuration owned by\n the affected capability.\n3. **Policy value** — it expresses what is allowed, limited, routed, retried,\n protected, or gated. Its typed definition belongs to the owning domain and it\n is evaluated through the policy system.\n4. **Derived value** — it is computed from other settings and is not\n independently mutable.\n5. **Runtime state** — it changes as the application operates, such as the\n active model for a conversation. It is changed by a domain command and is not\n configuration merely because it can be changed.\n\nAn unclassified behavioural literal is a defect.\n\n`const` MUST NOT be used merely because a value currently has one default.\nConstants are reserved for actual invariants: enum discriminants, protocol\nversions, fixed compatibility tokens, and values whose alteration would change\nthe program's semantics rather than configure an installation.\n\nIn Go, a constant is a compile-time value and does not represent mutable runtime\nstorage. It generally has no address that can be observed or changed.\nArchitecturally, a constant therefore asserts that the value cannot chaLine truncated
},
{
"url": "/dev/architecture/dashboard-agent-tools/",
"title": "The dashboard's tools for an in-browser agent",
"section": "Architecture",
"body": "# The dashboard's tools for an in-browser agent\n\nThe dashboard registers in-browser tools an agent can call through the page's model\ncontext. Four are read-only — `list_tasks`, `get_task`, `recent_events`,\n`daemon_health` — and three recover stuck work: `retry_task`, `stop_task`,\n`cancel_task`. Each wraps the same `ui/src/lib/api.ts` client the dashboard itself\ncalls, so an agent and a person exercise one path rather than two.\n\nThe four judgement verbs are withheld for the reason `identity.md` gives: an agent\nthat can approve can approve its own work.\n\n## What a tool call can be attributed to\n\nA page can invoke its own tools, so an agent's call and a page script's call reach\nthe server identically — the difference does not exist on the wire, and no header\nrecovers it. A tool call is therefore attributed to the signed-in identity the\nrequest presented, with its source recorded as a hint; it never records that an\nagent acted, and it never records that a human approved. An agent's action is\nattributable only on a surface where the agent presents its own credential.\n\n## Confirmation is a signal, not a control\n\nThe three recovery tools require an explicit `confirm: true`, which makes a\nconsequential call deliberate rather than incidental. It is not a security control:\nan agent can trivially satisfy it, and the browser cannot enforce\n`consequentialHint` at all — Chrome accepts it at registration and omits it from\n`getTools()`, so no client can see what the browser does not surface.\n\nThe boundary that holds is the session. These tools cannot exceed what the operator\nsigned in at the provider is already permitted to do.\n\n## Two ways the surface fails silently\n\n- **A secure context is required.** On a plain-HTTP origin `document.modelContext` is\n absent even with the flag enabled, and nothing reports why. `mkcert` covers local\n development.\n- **The dashboard embeds `ui/dist` into its binary.** Rebuilding the distribution is\n not deploying it: a running process can serve a bundle without tools while the\n built one has them, or serve a bundle older than its own binary.\n"
},
{
"url": "/dev/architecture/dependencies-and-contracts/",
"title": "Dependencies and Contracts",
@@ -46,7 +52,7 @@
"url": "/dev/architecture/identity/",
"title": "Identity",
"section": "Architecture",
"body": "# Identity\n\n**Status:** Approved foundation; lifecycle details are deferred \n**Date:** 2026-07-28 \n**Tracking issue:** [#73](https://github.com/samcharles93/archie-core/issues/73)\n\n## Purpose\n\nIdentity answers which persistent actor owns or performed an action. It supplies\nstable cross-domain attribution and lifecycle semantics.\n\nIdentity kinds include:\n\n- the Archie system identity, such as `System` or `archied`;\n- agent or bot identities;\n- service-account identities;\n- future user identities.\n\nA user is another kind of identity rather than a separate foundation concept.\n\n## Ownership\n\n`internal/domain/identity` owns:\n\n- immutable canonical `IdentityID` values;\n- identity kind;\n- lifecycle state;\n- ownership and attribution semantics;\n- identity-related commands and events.\n\nDomains may record an identity reference on their own state and events when\nownership or attribution is required. They MUST NOT receive a complete\nper-identity application configuration bundle.\n\nPersona and communication style are separate from identity.\n\nEvery Agent has its own immutable acting `IdentityID`. Agent actions use that\nidentity for ownership and attribution in the same way an organisation records\nwhich user accessed a resource, changed a permission, or performed another\nauditable action.\n\nAgent and Identity remain separate aggregates. Identity supplies the stable\nactor reference; Agent owns assistant behaviour, memory, specialisation,\nconversations, and capability context.\n\nAgent identities MUST NOT be shared between Agents. Automated WorkflowExecutions\nMAY eventually act through a dedicated service-account identity when a specific\nunattended automation requires it. Such use is an explicit execution decision\nand audit principal, not the default for Agents or a substitute for recording\nthe initiating user when one exists.\n\nA Workflow is owned by an IdentityID and executes as its owner by default.\nFuture delegated execution MAY explicitly assign a different `RunAsIdentityID`,\nincluding a user or service-account identity. Ownership, effective execution\nidentity, initiating identity, and delegation provenance remain distinct audit\nfacts.\n\n## Canonical identity\n\nEvery identity has an immutable `IdentityID`. It is the only value used for\ncross-domain ownership, attribution, routing, persistence, commands, and events.\n\nIdentity kind is explicit:\n\n- system;\n- bot;\n- service account;\n- user.\n\nDisplay names are mutable and MUST NOT be used as identifiers. Forge usernames,\nchannel bot names, Git authors, provider account names, and other external\nidentifiers are bindings associated with an identity rather than identity\nitself.\n\n`System` is represented by a real identity and `IdentityID`. The empty string\nMUST NOT retain special meaning as a legacy system identity.\n\nConfiguration refers to identities by `IdentityID` and supplies external\nbindings and capability settings separately. Renaming an identity or changing an\nexternal account MUST NOT change ownership or historical attribution.\n\nSecret-bearing channel bindings remain configuration concerns rather than\nidentity data. In particular, `[chat.telegram].token` is a\n`secret.SecretRef` resolved by the application through the configured secret\nregistry; the older `token_env` field remains a compatibility fallback. Neither\nform changes the identity that owns or performs a chat action.\n\n## Persistence\n\nIdentity records are durable domain state. The Identity domain's repository is\nthe source of truth; infrastructure implements its persistence contract.\n\nConfiguration MAY bootstrap a missing identity through an explicit, audited\ncreation operation and MAY propose runtime capability bindings for an existing\n`IdentityID`. Configuration is not the identity registry.\n\nOmitting an identity from configuration stops assembling the omitted runtime\ncapabilities. It MUST NOT delete the identity record, erase history, or make\nhistorical references unresolvable.\n\nDisabling or retiring an identity requires an explicit Identity-domain command.\nRetired identities remain resolvable for historical attribution. A failed\nexternal binding affects capability availability without destroying or silently\nreplacing the identity.\n\n## Lifecycle\n\nIdentity lifecycle is deliberately small:\n\n- `active`: may participate in runtime composition;\n- `suspended`: deliberately prevented from initiating new work while ownership\n and history remain;\n- `retired`: terminal, unable to initiate work, and permanently retained for\n attribution.\n\nAllowed transitions are:\n\n```text\ncreate -\u003e active\nactive -\u003e suspended\nsuspended -\u003e active\nactive -\u003e retired\nsuspended -\u003e retired\n```\n\n`retired` has no outgoing transition.\n\nCapability health is separate from identity lifecycle. Forge, model, channel,\nworkspace, and other bindings report health independently. An active ideLine truncated
"body": "# Identity\n\n**Status:** Approved foundation; lifecycle details are deferred \n**Date:** 2026-07-28 \n**Tracking issue:** [#73](https://github.com/samcharles93/archie-core/issues/73)\n\n## Purpose\n\nIdentity answers which persistent actor owns or performed an action. It supplies\nstable cross-domain attribution and lifecycle semantics.\n\nIdentity kinds include:\n\n- the Archie system identity, such as `System` or `archied`;\n- agent or bot identities;\n- service-account identities;\n- future user identities.\n\nA user is another kind of identity rather than a separate foundation concept.\n\n## Ownership\n\n`internal/domain/identity` owns:\n\n- immutable canonical `IdentityID` values;\n- identity kind;\n- lifecycle state;\n- ownership and attribution semantics;\n- identity-related commands and events.\n\nDomains may record an identity reference on their own state and events when\nownership or attribution is required. They MUST NOT receive a complete\nper-identity application configuration bundle.\n\nPersona and communication style are separate from identity.\n\nEvery Agent has its own immutable acting `IdentityID`. Agent actions use that\nidentity for ownership and attribution in the same way an organisation records\nwhich user accessed a resource, changed a permission, or performed another\nauditable action.\n\nAgent and Identity remain separate aggregates. Identity supplies the stable\nactor reference; Agent owns assistant behaviour, memory, specialisation,\nconversations, and capability context.\n\nAgent identities MUST NOT be shared between Agents. Automated WorkflowExecutions\nMAY eventually act through a dedicated service-account identity when a specific\nunattended automation requires it. Such use is an explicit execution decision\nand audit principal, not the default for Agents or a substitute for recording\nthe initiating user when one exists.\n\nA Workflow is owned by an IdentityID and executes as its owner by default.\nFuture delegated execution MAY explicitly assign a different `RunAsIdentityID`,\nincluding a user or service-account identity. Ownership, effective execution\nidentity, initiating identity, and delegation provenance remain distinct audit\nfacts.\n\n## Canonical identity\n\nEvery identity has an immutable `IdentityID`. It is the only value used for\ncross-domain ownership, attribution, routing, persistence, commands, and events.\n\nIdentity kind is explicit:\n\n- system;\n- bot;\n- service account;\n- user.\n\nDisplay names are mutable and MUST NOT be used as identifiers. Forge usernames,\nchannel bot names, Git authors, provider account names, and other external\nidentifiers are bindings associated with an identity rather than identity\nitself.\n\n`System` is represented by a real identity and `IdentityID`. The empty string\nMUST NOT retain special meaning as a legacy system identity.\n\nConfiguration refers to identities by `IdentityID` and supplies external\nbindings and capability settings separately. Renaming an identity or changing an\nexternal account MUST NOT change ownership or historical attribution.\n\nSecret-bearing channel bindings remain configuration concerns rather than\nidentity data. In particular, `[chat.telegram].token` is a\n`secret.SecretRef` resolved by the application through the configured secret\nregistry; the older `token_env` field remains a compatibility fallback. Neither\nform changes the identity that owns or performs a chat action.\n\n## Persistence\n\nIdentity records are durable domain state. The Identity domain's repository is\nthe source of truth; infrastructure implements its persistence contract.\n\nConfiguration MAY bootstrap a missing identity through an explicit, audited\ncreation operation and MAY propose runtime capability bindings for an existing\n`IdentityID`. Configuration is not the identity registry.\n\nOmitting an identity from configuration stops assembling the omitted runtime\ncapabilities. It MUST NOT delete the identity record, erase history, or make\nhistorical references unresolvable.\n\nDisabling or retiring an identity requires an explicit Identity-domain command.\nRetired identities remain resolvable for historical attribution. A failed\nexternal binding affects capability availability without destroying or silently\nreplacing the identity.\n\n## Lifecycle\n\nIdentity lifecycle is deliberately small:\n\n- `active`: may participate in runtime composition;\n- `suspended`: deliberately prevented from initiating new work while ownership\n and history remain;\n- `retired`: terminal, unable to initiate work, and permanently retained for\n attribution.\n\nAllowed transitions are:\n\n```text\ncreate -\u003e active\nactive -\u003e suspended\nsuspended -\u003e active\nactive -\u003e retired\nsuspended -\u003e retired\n```\n\n`retired` has no outgoing transition.\n\nCapability health is separate from identity lifecycle. Forge, model, channel,\nworkspace, and other bindings report health independently. An active ideLine truncated
},
{
"url": "/dev/architecture/messaging-and-work-intake/",
@@ -436,6 +442,10 @@
"title": "Configuration",
"url": "/dev/architecture/configuration/"
},
{
"title": "The dashboard's tools for an in-browser agent",
"url": "/dev/architecture/dashboard-agent-tools/"
},
{
"title": "Dependencies and Contracts",
"url": "/dev/architecture/dependencies-and-contracts/"
+1 -6
View File
@@ -9,10 +9,6 @@ Released 2026-09-22. [All news](../index.md)
`internal/app/controlplane`; the client half now lives in
`internal/infrastructure/controlplanerpc`, and the app package keeps the server, the registry, the
seeds, the validators and the schema derivation.
- **The PRD's guard now exists, and passes:** `cmd/archie-messaging/architecture_test.go` measures the
linked binary against the banned list, with `internal/domain/workflow/task` as an exception checked
in both directions. Before this the guard had never been written -- because it failed. A messaging
process linking a SQLite driver and the workflow engine is a severance hole, not a design choice.
- The resource documents are type-aliased back into the app package, so there is one definition read
under two names rather than a copy per side.
@@ -20,8 +16,7 @@ Released 2026-09-22. [All news](../index.md)
- The Messaging Service reports each channel's state, recorded **per channel**: one channel's failure
marks only itself, a channel that cannot start is marked failed **with its reason** rather than only
logged, and one that returns is marked stopped. `internal/channels/status.Manager` finally has a
production writer.
logged, and one that returns is marked stopped. `internal/channels/status.Manager` reports each channel's state.
- `PutChannelStatus`/`ListChannelStatus` carry that to the dashboard, following the config-snapshot
split. The store **replaces the reported set**, so a channel the reporter stops mentioning is
deleted rather than left as a stale `running`; an empty report clears the table, because hosting no
-23
View File
@@ -14,29 +14,6 @@ Released 2026-09-22. [All news](../index.md)
agent's call and a page script's call reach the server identically. Tools therefore record the
signed-in identity and a source, and **never** that an agent acted.
### The four judgement verbs are deliberately absent
- `approve`, `reject`, `abandon` and `archive` are **not** exposed. `internal/domain/taskactions`
records only a source string and stamps `KindHumanApproved`, so an agent's approval would be written
down as a person's. An agent that can approve can approve its own work, and the one control this
design exists to keep is a human accepting it. Those verbs land once task actions carry both the
acting identity and the authorising principal.
### `consequentialHint` does not gate anything, and the release says so
- Chrome 152 accepts `consequentialHint` at registration and **omits it from `getTools()`**, so no
client sees it and the browser cannot enforce what it does not surface. The three recovery tools
instead require an explicit `confirm: true`, which is a **deliberate-action signal rather than a
security control** an agent can trivially satisfy. The boundary that holds is the session: the tools
cannot exceed what the signed-in operator can already do.
### Two ways this fails silently, now recorded next to the code
- **A secure context is required.** On a plain-HTTP origin `document.modelContext` is absent *even
with the flag*, and nothing reports why. `mkcert` covers local development.
- **`archie-ui` embeds `ui/dist` into its binary**, so rebuilding the distribution is not deploying it;
a running process can serve a bundle with no tools while the built one has them.
### Binaries built by the development build task can be updated
- `task build` passed the `installtype` stamp for three of six commands, so `archie-messaging`,
+2 -2
View File
@@ -11,14 +11,14 @@
"name": "archied",
"version": "1.36.0",
"date": "2026-09-22",
"body": "### The dashboard declares tools an in-browser agent can call\n\n- The dashboard registers **seven WebMCP tools**. Read-only: `list_tasks`, `get_task`,\n `recent_events`, `daemon_health`. Recovery: `retry_task`, `stop_task`, `cancel_task`. An agent\n operating the dashboard now reads structured state and recovers stuck work instead of scraping the\n UI.\n- Every tool wraps the same `ui/src/lib/api.ts` client the dashboard itself calls, so an agent and a\n person exercise one path rather than two.\n- `archie-ui` cannot be attributed to the agent that called it: a page can invoke its own tools, so an\n agent's call and a page script's call reach the server identically. Tools therefore record the\n signed-in identity and a source, and **never** that an agent acted.\n\n### The four judgement verbs are deliberately absent\n\n- `approve`, `reject`, `abandon` and `archive` are **not** exposed. `internal/domain/taskactions`\n records only a source string and stamps `KindHumanApproved`, so an agent's approval would be written\n down as a person's. An agent that can approve can approve its own work, and the one control this\n design exists to keep is a human accepting it. Those verbs land once task actions carry both the\n acting identity and the authorising principal.\n\n### `consequentialHint` does not gate anything, and the release says so\n\n- Chrome 152 accepts `consequentialHint` at registration and **omits it from `getTools()`**, so no\n client sees it and the browser cannot enforce what it does not surface. The three recovery tools\n instead require an explicit `confirm: true`, which is a **deliberate-action signal rather than a\n security control** an agent can trivially satisfy. The boundary that holds is the session: the tools\n cannot exceed what the signed-in operator can already do.\n\n### Two ways this fails silently, now recorded next to the code\n\n- **A secure context is required.** On a plain-HTTP origin `document.modelContext` is absent *even\n with the flag*, and nothing reports why. `mkcert` covers local development.\n- **`archie-ui` embeds `ui/dist` into its binary**, so rebuilding the distribution is not deploying it;\n a running process can serve a bundle with no tools while the built one has them.\n\n### Binaries built by the development build task can be updated\n\n- `task build` passed the `installtype` stamp for three of six commands, so `archie-messaging`,\n `archie-state-store` and `archie-ui` were built without it. An unstamped binary reports `unknown`\n and the self-update path **refuses rather than guesses**, which is how an operator ends up unable to\n update with `install type is unknown` and no way forward. All six are stamped now, and a test reads\n the build task and fails naming any command line that omits it."
"body": "### The dashboard declares tools an in-browser agent can call\n\n- The dashboard registers **seven WebMCP tools**. Read-only: `list_tasks`, `get_task`,\n `recent_events`, `daemon_health`. Recovery: `retry_task`, `stop_task`, `cancel_task`. An agent\n operating the dashboard now reads structured state and recovers stuck work instead of scraping the\n UI.\n- Every tool wraps the same `ui/src/lib/api.ts` client the dashboard itself calls, so an agent and a\n person exercise one path rather than two.\n- `archie-ui` cannot be attributed to the agent that called it: a page can invoke its own tools, so an\n agent's call and a page script's call reach the server identically. Tools therefore record the\n signed-in identity and a source, and **never** that an agent acted.\n\n### Binaries built by the development build task can be updated\n\n- `task build` passed the `installtype` stamp for three of six commands, so `archie-messaging`,\n `archie-state-store` and `archie-ui` were built without it. An unstamped binary reports `unknown`\n and the self-update path **refuses rather than guesses**, which is how an operator ends up unable to\n update with `install type is unknown` and no way forward. All six are stamped now, and a test reads\n the build task and fails naming any command line that omits it."
},
{
"component": "archied",
"name": "archied",
"version": "1.35.0",
"date": "2026-09-22",
"body": "### The Messaging Service is severed from the store and workflow runtime\n\n- `cmd/archie-messaging` no longer links `modernc.org/sqlite`, `internal/store`,\n `internal/domain/workflow` or `internal/agentexec`. All four arrived through a single import of\n `internal/app/controlplane`; the client half now lives in\n `internal/infrastructure/controlplanerpc`, and the app package keeps the server, the registry, the\n seeds, the validators and the schema derivation.\n- **The PRD's guard now exists, and passes:** `cmd/archie-messaging/architecture_test.go` measures the\n linked binary against the banned list, with `internal/domain/workflow/task` as an exception checked\n in both directions. Before this the guard had never been written -- because it failed. A messaging\n process linking a SQLite driver and the workflow engine is a severance hole, not a design choice.\n- The resource documents are type-aliased back into the app package, so there is one definition read\n under two names rather than a copy per side.\n\n### Channel lifecycle reaches the dashboard\n\n- The Messaging Service reports each channel's state, recorded **per channel**: one channel's failure\n marks only itself, a channel that cannot start is marked failed **with its reason** rather than only\n logged, and one that returns is marked stopped. `internal/channels/status.Manager` finally has a\n production writer.\n- `PutChannelStatus`/`ListChannelStatus` carry that to the dashboard, following the config-snapshot\n split. The store **replaces the reported set**, so a channel the reporter stops mentioning is\n deleted rather than left as a stale `running`; an empty report clears the table, because hosting no\n channels is a fact worth reporting; and an id-less row is refused, since it could never be replaced\n or removed by a later report.\n- `Service.ReloadChannel` routes to the channel's own seam and refuses a channel that cannot reload\n with a reason rather than accepting and ignoring it. `ReloadSupported` was previously declared **by\n name**, so a channel could advertise reload with nothing behind it; one function now decides both.\n\n### Action playbooks: a second playbook shape, with typed results\n\n- A playbook is now one of two shapes: a **workflow playbook** (exactly one `workflow` action,\n unchanged) or an **action playbook** (one or more `module` actions in order). A playbook does not\n mix them.\n- `args` and `when` share one compilation path and one evaluation context, and `actions.<id>` is typed\n by the kind that action runs, so a misspelled result field, an unknown id, a map index, `in`, a\n comprehension, or a shape-foreign key (`kind` on a `workflow` action, `workflow` on a `module`\n action) is a **load failure** rather than a dispatch failure.\n- **Action playbooks load but do not run yet.** Routing arrives with the first side-effecting\n position; the daemon warns at load rather than accepting them silently.\n- The CEL environment is built per playbook, because result typing depends on the ids a playbook\n declares.\n- `ModuleRegistry.DecodeResult` turns a kind's flat map result into its typed `Result` before a later\n expression reads it. Without it a typed read evaluated against a shape the run never produces, and\n so silently read null."
"body": "### The Messaging Service is severed from the store and workflow runtime\n\n- `cmd/archie-messaging` no longer links `modernc.org/sqlite`, `internal/store`,\n `internal/domain/workflow` or `internal/agentexec`. All four arrived through a single import of\n `internal/app/controlplane`; the client half now lives in\n `internal/infrastructure/controlplanerpc`, and the app package keeps the server, the registry, the\n seeds, the validators and the schema derivation.\n- The resource documents are type-aliased back into the app package, so there is one definition read\n under two names rather than a copy per side.\n\n### Channel lifecycle reaches the dashboard\n\n- The Messaging Service reports each channel's state, recorded **per channel**: one channel's failure\n marks only itself, a channel that cannot start is marked failed **with its reason** rather than only\n logged, and one that returns is marked stopped. `internal/channels/status.Manager` reports each channel's state.\n- `PutChannelStatus`/`ListChannelStatus` carry that to the dashboard, following the config-snapshot\n split. The store **replaces the reported set**, so a channel the reporter stops mentioning is\n deleted rather than left as a stale `running`; an empty report clears the table, because hosting no\n channels is a fact worth reporting; and an id-less row is refused, since it could never be replaced\n or removed by a later report.\n- `Service.ReloadChannel` routes to the channel's own seam and refuses a channel that cannot reload\n with a reason rather than accepting and ignoring it. `ReloadSupported` was previously declared **by\n name**, so a channel could advertise reload with nothing behind it; one function now decides both.\n\n### Action playbooks: a second playbook shape, with typed results\n\n- A playbook is now one of two shapes: a **workflow playbook** (exactly one `workflow` action,\n unchanged) or an **action playbook** (one or more `module` actions in order). A playbook does not\n mix them.\n- `args` and `when` share one compilation path and one evaluation context, and `actions.<id>` is typed\n by the kind that action runs, so a misspelled result field, an unknown id, a map index, `in`, a\n comprehension, or a shape-foreign key (`kind` on a `workflow` action, `workflow` on a `module`\n action) is a **load failure** rather than a dispatch failure.\n- **Action playbooks load but do not run yet.** Routing arrives with the first side-effecting\n position; the daemon warns at load rather than accepting them silently.\n- The CEL environment is built per playbook, because result typing depends on the ids a playbook\n declares.\n- `ModuleRegistry.DecodeResult` turns a kind's flat map result into its typed `Result` before a later\n expression reads it. Without it a typed read evaluated against a shape the run never produces, and\n so silently read null."
},
{
"component": "archied",