@arbiterforge/ca-pi 0.18.4 → 0.20.0

package/plugins/ca-pi/agents/infra-author.md

@@ -9,12 +9,11 @@ model: sonnet
 
 # Infrastructure Author Agent
 
-Infrastructure implementation executor. Write IaC, container, CI/CD, and deployment
-configuration only after the owning route clears its prerequisite: approved
-feature criteria (and a plan where applicable) for `/feature`, or the original
-caller's repair authority and a bug-origin regression obligation for `/fix`.
-The regression must be red for the
-right reason before implementation. Do not fabricate a feature artifact for a fix.
+Infrastructure implementation executor. Use the caller's scoped obligations and
+selected verification approach: agreed feature criteria, an authorized bug-origin
+regression obligation, or the refactor's approved surface and parity contract.
+For `/fix`, observe target RED before repair. Do not fabricate a feature artifact
+for a fix. Return scoped results; stage only when assigned and authorized.
 
 ## Required Reading at the Start of Every Task
 
@@ -22,7 +21,7 @@ Read in full before writing any infrastructure code:
 
 1. `<project-root>/.codearbiter/tech-stack.md` — IaC tool (OpenTofu, Ansible, Helm, etc.), container runtime, CI/CD system, approved registries
 2. `<project-root>/.codearbiter/security-controls.md` — security boundaries, approved secrets store, compliance requirements affecting infrastructure
-3. `<plugin-root>/includes/author-tdd-workflow.md` — the shared six-step TDD execution order. Read it; do not carry a remembered copy.
+3. `<plugin-root>/includes/author-tdd-workflow.md` — route-specific verification and the authorized handoff.
 
 ## Security Rules
 

package/plugins/ca-pi/agents/scout.md

@@ -18,28 +18,24 @@ Dispatched by `decision-variance` and `context-creation` to scan an assigned cod
 - Codebase sections have natural ownership boundaries that map to scout assignments.
 - Parallel evidence gathering across multiple areas is needed.
 
-Under ~50 files, `decision-variance` scans inline. When a qualified context
-profile becomes available, `context-creation` dispatches its six fixed,
-isolated scout assignments because its report-only synthesis boundary depends
-on the orchestrator not reading raw source.
+Under ~50 files, `decision-variance` scans inline. `context-creation` uses
+direct bounded inspection or optional scoped agents according to the work.
+Six logical categories do not require six children or parallel execution.
 
 ## Context-creation assignment
 
-The shared `scout` frontmatter includes Bash for existing `decision-variance`
-assignments. It is therefore not a context-creation containment profile.
-`context-creation` must BLOCK before dispatch until the active host supplies a
-separate, fresh child identity with host-enforced read-only restriction and an
-observed denied benign write, successful source read, and bounded report
-delivery. A prose instruction to avoid writes is not enforcement. An explicit
-session-only `context-scout` may retain Bash solely for exact, frozen
-`git hash-object -- '<source path>'` reads when a caller-owned command guard and
-read-only OS phase are independently qualified. No other Bash command is part
-of this assignment, and the inherited shared `scout` role remains ineligible. This
-assignment preserves report-only synthesis; the orchestrator uses reports and
-does not load raw source after pre-flight. A qualified context scout must emit
-the versioned closed report envelope accepted by `_contextreportlib`, bound to
-its assignment, attempt, snapshot and scope. A trusted read-only hash collector
-must supply each cited digest when the restricted child has no Bash.
+Read only the assigned source scope and return a bounded evidence report.
+Respect actual host permissions, excluded paths and secret-handling rules.
+The role's tools and read-only instructions are not evidence of containment;
+do not claim a trusted sandbox, independent verification or authority from
+this role label. A separately required host restriction remains binding.
+Do not modify files, execute repository scripts or spawn further agents.
+
+Name the assignment, inspected paths and source identity, citations/digests,
+coverage, unknowns and conflicts. The parent can directly inspect a necessary
+source or resolve a missing report. Use `_contextreportlib`'s closed envelope
+when the caller selects that structured transport; do not invent an envelope
+or receipt to make direct inspection appear delegated.
 
 For context-creation, report each assigned category even when the search finds
 no match. Use the closed outcome vocabulary: `observed`, `not_applicable`,
@@ -83,7 +79,13 @@ Constraints:
 
 Decision-variance scouts return this exact format.
 
-**Hash field:** For decision-variance assignments, emit each cited file's `git hash-object <path>` oid in the `hash` field. That legacy role has Bash. A context-creation assignment needs a digest in the method and shape required by the closed report decoder. A qualified context-specific Bash route may use only its frozen exact `git hash-object -- '<source path>'` commands and must label the result `git-blob-normalized-sha1`. A different qualified read-only collector may provide `sha256-raw`, retaining that distinct method identity. A missing actual digest blocks the citation; neither the caller's snapshot checksum nor an injected answer substitutes for the child's or trusted collector's observed result. A git oid is not a code excerpt; the "no excerpts" rule (file path and line numbers only) is not violated by including an oid.
+**Hash field:** For decision-variance assignments, emit each cited file's
+`git hash-object <path>` oid in the `hash` field. Context assignments retain
+the actual collector and digest method, distinguishing `git-blob-normalized-sha1`
+from `sha256-raw`. Do not substitute a snapshot checksum for a cited file's
+observed digest. A source hash binds bytes, not semantic correctness. Missing
+digests remain unknown until a permitted read obtains them; no secret values
+or raw excerpts belong in the report.
 
 ```markdown
 # Scout Report — <scope>
@@ -120,9 +122,8 @@ The scout reports file path and line number only — no code excerpts, quoted te
 1. **Signal-to-noise** — excerpts grow unbounded under load; entire files get pasted "for context" and reports become unreadable.
 2. **Sensitive content** — scaffold may carry secrets, internal URLs, cloud account IDs, credentials, hostnames. Quoting those leaks them into the dispatching skill's working files.
 
-When `decision-variance` needs actual content, it may read the file directly.
-`context-creation` never does so after Phase 1; it synthesizes only from the six
-scout reports.
+When either owning workflow needs actual content, it may read the necessary
+source directly under the same host permissions and secret-handling boundaries.
 
 ## Scout Confidence Levels
 
@@ -138,7 +139,7 @@ The scout MUST NOT:
 
 1. **Judge correctness.** "This looks wrong" is not the scout's call.
 2. **Compare scope against artifacts.** The scout has no knowledge of them.
-3. **Speculate beyond evidence.** No evidence in a file → report it absent.
+3. **Speculate beyond evidence.** No match → report the searched scope and limits, not universal absence.
 4. **Modify files.** Read-only.
 5. **Spawn further subagents.** Only the dispatching skill spawns subagents.
 6. **Include excerpts, quotes, or pasted content.** File path and line number only.

package/plugins/ca-pi/arbiter.md

@@ -27,7 +27,7 @@ no emojis, no flattery.
 - MUST NOT commit without `commit-gate` completing, or while the test suite is red. Sole exception: a `spike/*` branch (via `/spike`), which can never merge or PR.
 - MUST NOT resolve a `[CONFIRM-NN]` by guessing.
 - MUST NOT redefine domain vocabulary without updating `.codearbiter/CONTEXT.md`.
-- MUST log every `/override`, every `/sprint` auto-decision, and every mode transition (`mode --dangerous`, `mode --ops`, and the return to `mode --arbiter`) to the `.codearbiter/` audit trail.
+- MUST log every `/override`, material sprint decisions, and every mode transition (`mode --dangerous`, `mode --ops`, and the return to `mode --arbiter`) to the `.codearbiter/` audit trail.
 - MUST load skill/routine bodies on invocation only; the `INDEX.md` files are the surface scan. No bulk reads.
 
 ---
@@ -61,9 +61,11 @@ Before dispatching roles, editing audit files, or using native compaction, load
 
 **Loaded fully on invocation, never acted on from memory:**
 
-- `/ca-sprint` — autonomous sprint: load and follow `<plugin-root>/SPRINT.md`. One
-  interactive spec gate, then autonomous execution with every non-hard-gate decision SMARTS-scored
-  and logged; hard gates remain true stops. A trailing `--farm` flag passes through to `SPRINT.md`.
+- `/ca-sprint` — autonomous sprint: load and follow `<plugin-root>/SPRINT.md`. Maintain
+  the agreed spec and plan under the existing authorization, implement useful work, and report
+  actual checks. Ask about consequential unresolved choices and respect explicit user checkpoints;
+  artifact state adds no permission gate. Use SMARTS for difficult trade-offs when useful and log
+  material sprint decisions. A trailing `--farm` flag passes through to `SPRINT.md`.
 
 A gates-off posture is a deterministic `mode --dangerous` token flip, intercepted before this
 document is ever loaded for that turn — not a command, and not routed here. `includes/dangerous-mode.md`

package/plugins/ca-pi/generated/command-catalog.json

@@ -143,7 +143,7 @@
     },
     "context-check": {
       "name": "context-check",
-      "description": "Audit stale provenance-tracked docs on request. Report first; re-scout or re-baseline only for selected docs.",
+      "description": "Inspect context drift and refresh selected source-backed documents under existing authorization.",
       "skillPath": "skills/ca-context-check/SKILL.md",
       "visibility": "alias",
       "workflow": "operate",
@@ -152,7 +152,7 @@
     },
     "create-context": {
       "name": "create-context",
-      "description": "Build project context from an existing codebase through isolated scouts, resolve gaps, and preserve initialization gates.",
+      "description": "Build source-backed project context through bounded inspection, optional agents, and safe selected writes.",
       "skillPath": "skills/ca-create-context/SKILL.md",
       "visibility": "alias",
       "workflow": "initialize",
@@ -170,7 +170,7 @@
     },
     "decompose": {
       "name": "decompose",
-      "description": "Develop greenfield project context through a layered interview, preserve decisions, and initialize only after the required gates.",
+      "description": "Develop greenfield context from a durable interview and initialize it through the native context writer.",
       "skillPath": "skills/ca-decompose/SKILL.md",
       "visibility": "alias",
       "workflow": "initialize",
@@ -188,7 +188,7 @@
     },
     "feature": {
       "name": "feature",
-      "description": "Start a feature: brainstorm a spec, get it approved, then drive it test-first through the pipeline. The one entry to implementation.",
+      "description": "Build an authorized feature from a maintained specification, testable criteria and useful plan, without duplicating user approval.",
       "skillPath": "skills/ca-feature/SKILL.md",
       "visibility": "core",
       "workflow": "change",
@@ -311,7 +311,7 @@
     },
     "sprint": {
       "name": "sprint",
-      "description": "Autonomous sprint — one interactive spec gate, then plan-to-PR execution with every auto-decision SMARTS-scored and logged. Hard gates remain true stops.",
+      "description": "Run an autonomous sprint from maintained specs and plans, with truthful verification and no duplicate artifact approval ceremonies.",
       "skillPath": "skills/ca-sprint/SKILL.md",
       "visibility": "core",
       "workflow": "change",

package/plugins/ca-pi/helpers/artifacts/release.json

@@ -2,47 +2,47 @@
   "format": "codearbiter.artifact-release/0.1.0",
   "version": "0.1.0",
   "protocol": "codearbiter.artifact-api/0.1.0",
-  "schema_version": "0.3.1",
+  "schema_version": "0.3.2",
   "binaries": {
     "darwin/amd64": {
       "file": "ca-artifact-darwin-amd64",
-      "sha256": "42f302c05c66b22294bff0bdacbd28d5373ddd64b447f087a4532cd99f0a1b6c",
+      "sha256": "ca6dc89a106f4b79eb2ab992eb7617c79742f88eb3d5d9b0444102ecf3f23ed3",
       "native_tested": true
     },
     "darwin/arm64": {
       "file": "ca-artifact-darwin-arm64",
-      "sha256": "adf818a92d6d45db8d13a4ec685b2e2e14ddd19d0978c3d820541c45ed5cd4d9",
+      "sha256": "04dae5874b003fb5530a460c3d9ce6da8284ca7bbfb886e6b6591d0e2980934d",
       "native_tested": true
     },
     "linux/amd64": {
       "file": "ca-artifact-linux-amd64",
-      "sha256": "8090a552d8a40bd3d37075eb81005d4e2bc4cdefac3ecb04cef8d72ef00f64d5",
+      "sha256": "1b2d858cf64e7774cbc2abf8abe7921c5f9fe147fcee796e70dbe38478272847",
       "native_tested": true
     },
     "linux/arm64": {
       "file": "ca-artifact-linux-arm64",
-      "sha256": "0e65ced478bc490e80d6ab5323d307ceffdf42830709563cd6740ef5f35fb069",
+      "sha256": "37e592fb61297850b17e2cd6c603a203143940b42b04e7a285f081b0d9d848e5",
       "native_tested": true
     },
     "windows/amd64": {
       "file": "ca-artifact-windows-amd64.exe",
-      "sha256": "687e362c216634b531371b4238d129a5f09df716d8cf73b6123f20d4c40c9d36",
+      "sha256": "4f4b3f4cf0ab7230f9cd577a3310a293007af1126913224c8e6103438aaafc8b",
       "native_tested": true
     },
     "windows/arm64": {
       "file": "ca-artifact-windows-arm64.exe",
-      "sha256": "bc4920311a4e3768439b9c35bd0aca77f1ab964011d7812dd8932fc7360527a0",
+      "sha256": "855bfbd1c628a0b76cfa5bdf7d6ff392fb55e49dba23efb13e2a7a26d9e5ccb4",
       "native_tested": true
     }
   },
   "context_qualification": {
     "darwin/amd64": {
-      "source_commit": "045c0812888f20f4c843677cfb37b802e2084770",
+      "source_commit": "44149055d1ca136e191f79136244f0d3a2b1a461",
       "workflow": ".github/workflows/ci.yml",
-      "run_id": "38575955363",
-      "binary_sha256": "42f302c05c66b22294bff0bdacbd28d5373ddd64b447f087a4532cd99f0a1b6c",
-      "observation_sha256": "e03881ae6e4baf8095597869933ae43dd90b4e1c97bd56dbd83ec57535f6ad7f",
-      "module_sha256": "e482ad7b0c9264711828ac8404c624517b893981b45712d70ead0fb0ef3b5522",
+      "run_id": "39097065693",
+      "binary_sha256": "ca6dc89a106f4b79eb2ab992eb7617c79742f88eb3d5d9b0444102ecf3f23ed3",
+      "observation_sha256": "178ac16ede5052a25eff69597ef2c6ab30439bcfd3d9127a1189e9eaa7a6147e",
+      "module_sha256": "a34f09f52c52145624e87c1d5e3850c822dd1ca54c346c5ad63802f6e5a0d0b3",
       "native_cases": [
         "TestContextBoundedRender",
         "TestContextDocumentContract",
@@ -53,12 +53,12 @@
       ]
     },
     "darwin/arm64": {
-      "source_commit": "045c0812888f20f4c843677cfb37b802e2084770",
+      "source_commit": "44149055d1ca136e191f79136244f0d3a2b1a461",
       "workflow": ".github/workflows/ci.yml",
-      "run_id": "38575955363",
-      "binary_sha256": "adf818a92d6d45db8d13a4ec685b2e2e14ddd19d0978c3d820541c45ed5cd4d9",
-      "observation_sha256": "053780b4155a329003e9d8402a90be022ac1750e5f1420ea46a5079931464fd6",
-      "module_sha256": "e482ad7b0c9264711828ac8404c624517b893981b45712d70ead0fb0ef3b5522",
+      "run_id": "39097065693",
+      "binary_sha256": "04dae5874b003fb5530a460c3d9ce6da8284ca7bbfb886e6b6591d0e2980934d",
+      "observation_sha256": "c7a246aafb77097d60704ecae57abcd7a8786078144de3549475684cd93c1fc2",
+      "module_sha256": "a34f09f52c52145624e87c1d5e3850c822dd1ca54c346c5ad63802f6e5a0d0b3",
       "native_cases": [
         "TestContextBoundedRender",
         "TestContextDocumentContract",
@@ -69,12 +69,12 @@
       ]
     },
     "linux/amd64": {
-      "source_commit": "045c0812888f20f4c843677cfb37b802e2084770",
+      "source_commit": "44149055d1ca136e191f79136244f0d3a2b1a461",
       "workflow": ".github/workflows/ci.yml",
-      "run_id": "38575955363",
-      "binary_sha256": "8090a552d8a40bd3d37075eb81005d4e2bc4cdefac3ecb04cef8d72ef00f64d5",
-      "observation_sha256": "e83da7b877c8e37761b826b6a45bee44e009066a66c18309d2f854ed0d94096b",
-      "module_sha256": "e482ad7b0c9264711828ac8404c624517b893981b45712d70ead0fb0ef3b5522",
+      "run_id": "39097065693",
+      "binary_sha256": "1b2d858cf64e7774cbc2abf8abe7921c5f9fe147fcee796e70dbe38478272847",
+      "observation_sha256": "f2725ba4c6ef48d36fb92962ef1fc6b00871e057a3234010ab0066a844bf21fd",
+      "module_sha256": "a34f09f52c52145624e87c1d5e3850c822dd1ca54c346c5ad63802f6e5a0d0b3",
       "native_cases": [
         "TestContextBoundedRender",
         "TestContextDocumentContract",
@@ -85,12 +85,12 @@
       ]
     },
     "linux/arm64": {
-      "source_commit": "045c0812888f20f4c843677cfb37b802e2084770",
+      "source_commit": "44149055d1ca136e191f79136244f0d3a2b1a461",
       "workflow": ".github/workflows/ci.yml",
-      "run_id": "38575955363",
-      "binary_sha256": "0e65ced478bc490e80d6ab5323d307ceffdf42830709563cd6740ef5f35fb069",
-      "observation_sha256": "dbaace14fca6456b29dbb71360fe9d2f0ab73738bad9c09486f0964358958a07",
-      "module_sha256": "e482ad7b0c9264711828ac8404c624517b893981b45712d70ead0fb0ef3b5522",
+      "run_id": "39097065693",
+      "binary_sha256": "37e592fb61297850b17e2cd6c603a203143940b42b04e7a285f081b0d9d848e5",
+      "observation_sha256": "3c0cae7073c8e7cf7ba6e40da4a04fad9d623a87914e44dac3b6e329d0dd020a",
+      "module_sha256": "a34f09f52c52145624e87c1d5e3850c822dd1ca54c346c5ad63802f6e5a0d0b3",
       "native_cases": [
         "TestContextBoundedRender",
         "TestContextDocumentContract",
@@ -101,12 +101,12 @@
       ]
     },
     "windows/amd64": {
-      "source_commit": "045c0812888f20f4c843677cfb37b802e2084770",
+      "source_commit": "44149055d1ca136e191f79136244f0d3a2b1a461",
       "workflow": ".github/workflows/ci.yml",
-      "run_id": "38575955363",
-      "binary_sha256": "687e362c216634b531371b4238d129a5f09df716d8cf73b6123f20d4c40c9d36",
-      "observation_sha256": "575df189640f478ffdbc593f4a070c4e407b0f50b7787d20632eb3586bc16cee",
-      "module_sha256": "e482ad7b0c9264711828ac8404c624517b893981b45712d70ead0fb0ef3b5522",
+      "run_id": "39097065693",
+      "binary_sha256": "4f4b3f4cf0ab7230f9cd577a3310a293007af1126913224c8e6103438aaafc8b",
+      "observation_sha256": "3ad94468cbd6eaf40500b525859923feaa5fbb120f91b3aadcc08b08ae837fc9",
+      "module_sha256": "a34f09f52c52145624e87c1d5e3850c822dd1ca54c346c5ad63802f6e5a0d0b3",
       "native_cases": [
         "TestContextBoundedRender",
         "TestContextDocumentContract",
@@ -117,12 +117,12 @@
       ]
     },
     "windows/arm64": {
-      "source_commit": "045c0812888f20f4c843677cfb37b802e2084770",
+      "source_commit": "44149055d1ca136e191f79136244f0d3a2b1a461",
       "workflow": ".github/workflows/ci.yml",
-      "run_id": "38575955363",
-      "binary_sha256": "bc4920311a4e3768439b9c35bd0aca77f1ab964011d7812dd8932fc7360527a0",
-      "observation_sha256": "1b6574772d1e93c2cd59ff43d519e343d44926312cf4d90b840207b0a625d3f9",
-      "module_sha256": "e482ad7b0c9264711828ac8404c624517b893981b45712d70ead0fb0ef3b5522",
+      "run_id": "39097065693",
+      "binary_sha256": "855bfbd1c628a0b76cfa5bdf7d6ff392fb55e49dba23efb13e2a7a26d9e5ccb4",
+      "observation_sha256": "8f594d61907c66818ddae106e859c6ba437713ca4c049468425558751ddb9697",
+      "module_sha256": "a34f09f52c52145624e87c1d5e3850c822dd1ca54c346c5ad63802f6e5a0d0b3",
       "native_cases": [
         "TestContextBoundedRender",
         "TestContextDocumentContract",

package/plugins/ca-pi/hooks/_approvallib.py

@@ -345,6 +345,38 @@ def cancel_user_approval(root: str | Path, artifact_id: str) -> dict[str, Any]:
     return {"artifact_id": artifact_id, "cancelled": True}
 
 
+def retire_unanswered_user_approval(
+    root: str | Path, artifact_id: str, *, reason: str,
+) -> dict[str, Any]:
+    """Retire only an unsubmitted historical paired prompt, not an approval.
+
+    The older single-artifact 0.1 marker cannot establish whether a reply
+    was captured and its write outcome lost. It is therefore not eligible
+    for this automatic cleanup. The paired adapter serializes cancellation
+    against the genuine reply consumer and checks the durable submission.
+    """
+    if (not isinstance(artifact_id, str) or re.fullmatch(r"[A-Z][A-Z0-9-]{1,79}", artifact_id) is None
+            or not isinstance(reason, str) or not reason.strip()
+            or len(reason.encode("utf-8")) > 2048):
+        raise ApprovalError("INVALID_RETIREMENT", "retirement requires a valid artifact ID and bounded nonblank reason")
+    root = Path(root).resolve(strict=True)
+    pending = _load_pending(root)
+    if pending is None or pending["artifact_id"] != artifact_id:
+        raise ApprovalError("NO_PENDING_APPROVAL", "no matching pending approval request")
+    if pending["format"] != "codearbiter.pending-user-approval/0.2.0":
+        raise ApprovalError("LEGACY_APPROVAL_RECOVERY_REQUIRED",
+                            "single-artifact pending state cannot prove a user reply was unsubmitted")
+    import _sprintapprovallib
+    # Uses the existing private per-repository lock and independently checks
+    # the current marker and durable submission before deleting either route.
+    _sprintapprovallib.cancel(root, pending)
+    return {
+        "artifact_id": artifact_id, "disposition": "unanswered_prompt_superseded",
+        "reason": reason.strip(), "authority_granted": False, "mutation_attempted": False,
+    }
+
+
+
 def consume_user_approval(
     root: str | Path,
     client: Any,
@@ -489,6 +521,10 @@ def main(argv: list[str] | None = None) -> int:
     cancel = sub.add_parser("cancel")
     cancel.add_argument("--root", required=True)
     cancel.add_argument("--artifact-id", required=True)
+    retire = sub.add_parser("retire")
+    retire.add_argument("--root", required=True)
+    retire.add_argument("--artifact-id", required=True)
+    retire.add_argument("--reason", required=True)
     pair = sub.add_parser("arm-sprint")
     pair.add_argument("--root", required=True)
     pair.add_argument("--spec-id", required=True)
@@ -504,6 +540,11 @@ def main(argv: list[str] | None = None) -> int:
     if args.command == "cancel":
         print(json.dumps(cancel_user_approval(args.root, args.artifact_id)))
         return 0
+    if args.command == "retire":
+        print(json.dumps(retire_unanswered_user_approval(
+            args.root, args.artifact_id, reason=args.reason,
+        )))
+        return 0
     client = _artifactlib.ArtifactClient(
         args.root, _artifactlib.helper_installation(__file__)
     )