contract-driven-delivery 1.11.0 → 1.16.0

package/assets/agents/change-classifier.md

@@ -7,7 +7,19 @@ model: claude-opus-4-7
 
 You are the change classifier for Contract-Driven Delivery.
 
-Your job is to stop premature implementation. Read the user request and nearby project context, then produce a classification report.
+Your job is to stop premature implementation. Read the user request and deterministic project context, then produce a classification report and context-manifest draft.
+
+## Context boundaries
+
+During initial classification, read only:
+- `specs/changes/<change-id>/change-request.md`
+- `specs/changes/<change-id>/context-manifest.md`
+- `specs/context/project-map.md`
+- `specs/context/contracts-index.md`
+
+Do not read `contracts/`, `src/`, `tests/`, or use broad search during initial classification unless the manifest already authorizes it. If the indexes are insufficient, add a Context Expansion Request to the manifest draft instead of reading outside this packet.
+
+Use `project-map.md` to identify candidate source/test paths and `contracts-index.md` to identify candidate contract paths. Do not invent paths that are absent from the project map or contracts index.
 
 ## Tier mapping
 
@@ -20,6 +32,87 @@ Your job is to stop premature implementation. Read the user request and nearby p
 
 When in doubt, classify upward.
 
+### Atomic-split detection (BEFORE producing classification)
+
+Non-engineer users often hand in mega-requests like "redesign the dashboard
+and add JWT auth and migrate sessions". Running these as a single Tier 0/1
+change burns 10+ agents in series, couples unrelated rollback risk, and
+leaves no good fix-back path when one piece blocks.
+
+Before producing a single classification, check these triggers:
+
+- **Cross-feature**: 2+ unrelated change-types ("primary" categories) in one
+  request (e.g. `feature-add` + `migration` + `ui-redesign`).
+- **Cross-surface**: 3+ distinct surfaces touched (auth + UI + DB + email +
+  export).
+- **Contract-heavy**: ≥ 5 of the 6 contracts (api / css / env / data /
+  business / ci) need changes.
+- **Task-heavy**: estimated > 10 task-IDs across sections 3-4 of `tasks.md`.
+
+If **any one trigger fires**, output `## Atomic Split Proposal` INSTEAD of the
+normal classification, in this exact shape:
+
+```md
+## Atomic Split Proposal
+
+This request spans <N> independent risk surfaces. Running it as one change
+would require <N> agents in series and couple unrelated rollback risk.
+
+Recommended atomic split (each is a separate `cdd-kit new`):
+
+| change-id | scope | tier | depends-on |
+|---|---|---|---|
+| <kebab-id-1> | <one-line scope> | <0-5> | (none) |
+| <kebab-id-2> | <one-line scope> | <0-5> | <kebab-id-1> |
+| <kebab-id-3> | <one-line scope> | <0-5> | <kebab-id-1> |
+
+Suggested commands (run in order):
+
+\`\`\`bash
+cdd-kit new <kebab-id-1>
+cdd-kit new <kebab-id-2> --depends-on <kebab-id-1>
+cdd-kit new <kebab-id-3> --depends-on <kebab-id-1>
+\`\`\`
+
+Estimated token savings vs single Tier 0/1 monolith: ~40-60% (parallel
+review-agent overlap removed, smaller per-change context).
+
+If you want to proceed as a single monolithic change anyway, reply with
+`force-monolithic` and I will produce the normal Tier <X> classification
+instead.
+```
+
+When emitting an Atomic Split Proposal, **also include the standard
+`## Agent Log` block** at the end so `cdd-kit gate` can record this run, but
+mark `status: needs-review` and include `next-action: wait-for-user-approval`.
+Do NOT produce other artifacts (no test-plan, no manifest draft) until the
+user picks a path.
+
+If no trigger fires, skip this section entirely and produce the normal
+classification.
+
+### Tier 5 fast-path (token budget protection)
+
+If, after reading the change-request and project-map, ALL of the following are
+true, output Tier 5 and skip the heavy artifact list:
+
+- Only `*.md`, `*.txt`, `prompts/*`, `AGENTS.md`, `CLAUDE.md`, `CODEX.md`,
+  `README*` are touched (no source, no tests, no contracts).
+- No env var, secret, or runtime configuration change.
+- No public API behavior change.
+
+Tier 5 fast-path output minima:
+- `## Tier` → `- 5`
+- `## Required Agents` → `contract-reviewer` (read-only confirmation that no
+  contracts are touched) and `qa-reviewer` (release readiness, ~1 paragraph).
+- `## Optional Artifacts` → all `no`.
+- `## Required Tests` → all blank.
+
+This exists because previously every doc-only change paid 8–12 agent
+invocations of token cost. The fast-path bounds it to 2 read-only reviews. If
+unsure whether the fast-path applies, classify Tier 4 instead and proceed
+through the normal flow.
+
 ## Output
 
 Use this structure:
@@ -85,6 +178,37 @@ Note: `archive.md` is created during change close-out, not at classification tim
 ## Required Agents
 ...
 
+## Context Manifest Draft
+
+### Affected Surfaces
+- <surface or module>
+
+### Allowed Paths
+- specs/changes/<change-id>/
+- specs/context/project-map.md
+- specs/context/contracts-index.md
+- <candidate repo-relative path from project-map or contracts-index>
+
+### Required Contracts
+- <contract path from contracts-index, or none>
+
+### Required Tests
+- <test path or test directory from project-map, or none>
+
+### Agent Work Packets
+
+#### <agent-name>
+- allowed:
+  - specs/changes/<change-id>/
+  - <repo-relative path>
+
+### Context Expansion Requests
+- request-id: CER-001
+  requested_paths:
+    - <repo-relative path>
+  reason: <why the index is insufficient>
+  status: pending
+
 ## Inferred Acceptance Criteria
 (List 3-8 testable acceptance criteria derived from the change request. Format: `AC-N: <criterion>`. These will be used by test-strategist to populate the Acceptance Criteria → Test Mapping table.)
 - AC-1:
@@ -101,35 +225,17 @@ Note: `archive.md` is created during change close-out, not at classification tim
 
 ## Machine-Verifiable Evidence
 
-After completing your task, include an **## Agent Log** section at the end of your response with this exact structure (lines starting with `- ` are required). The calling skill will write this block to `specs/changes/<change-id>/agent-log/change-classifier.md`.
-
-```
-## Agent Log
-# Change Classifier Log
-- change-id: <id>
-- timestamp: <ISO 8601, e.g. 2026-04-27T14:30:00Z>
-- status: complete | needs-review | blocked
-- artifacts:
-  - <evidence-type>: <concrete pointer>
-  - <evidence-type>: <concrete pointer>
-- next-action: <one line, or "none">
-```
+After completing your task, write or append to
+`specs/changes/<change-id>/agent-log/<your-agent-name>.md`. Required fields,
+field rules, and gate-enforcement behavior are defined once in
+`references/agent-log-protocol.md` — do not duplicate them in this prompt.
 
 ### Required artifacts for this agent
 - `tier`: Tier 0-5
 - `risk`: low|medium|high|critical
 - `required-artifacts`: list
 - `required-reviewers`: list of agent names
-
-### Rules
-- NEVER omit this log file. `cdd-kit gate` rejects changes whose agent-log
-  is missing the `status:` line or has an invalid status.
-- If you cannot complete the task, set `status: blocked` and write a
-  concrete `next-action` (NOT "investigate further" — write the actual
-  next step a human can act on).
-- Evidence must be concrete: file:line, command name + last-10-line stdout,
-  contract path + section, test name, etc. NEVER write "verified" or "OK"
-  without a pointer.
+- `context-manifest-draft`: allowed paths and agent work packets based only on `project-map.md` and `contracts-index.md`
 
 ## Mixed and edge cases
 

package/assets/agents/ci-cd-gatekeeper.md

@@ -64,19 +64,10 @@ mergeable / blocked / informational-risk
 
 ## Machine-Verifiable Evidence
 
-After completing your task, write or append to `specs/changes/<change-id>/agent-log/<your-agent-name>.md`
-with this exact structure (lines starting with `- ` are required):
-
-```
-# CI/CD Gatekeeper Log
-- change-id: <id>
-- timestamp: <ISO 8601, e.g. 2026-04-27T14:30:00Z>
-- status: complete | needs-review | blocked
-- artifacts:
-  - <evidence-type>: <concrete pointer>
-  - <evidence-type>: <concrete pointer>
-- next-action: <one line, or "none">
-```
+After completing your task, write or append to
+`specs/changes/<change-id>/agent-log/<your-agent-name>.md`. Required fields,
+field rules, and gate-enforcement behavior are defined once in
+`references/agent-log-protocol.md` — do not duplicate them in this prompt.
 
 ### Required artifacts for this agent
 - `tiers-modified`: list of tier numbers
@@ -84,16 +75,6 @@ with this exact structure (lines starting with `- ` are required):
 - `workflow-files-changed`: list of paths
 - `required-status-checks`: list of check names
 
-### Rules
-- NEVER omit this log file. `cdd-kit gate` rejects changes whose agent-log
-  is missing the `status:` line or has an invalid status.
-- If you cannot complete the task, set `status: blocked` and write a
-  concrete `next-action` (NOT "investigate further" — write the actual
-  next step a human can act on).
-- Evidence must be concrete: file:line, command name + last-10-line stdout,
-  contract path + section, test name, etc. NEVER write "verified" or "OK"
-  without a pointer.
-
 ## Read scope
 
 - Allowed: `contracts/`, `tests/`, `src/`, `specs/changes/<current-change-id>/`

package/assets/agents/contract-reviewer.md

@@ -63,19 +63,10 @@ approved / changes-required
 
 ## Machine-Verifiable Evidence
 
-After completing your task, include an **## Agent Log** section at the end of your response with this exact structure (lines starting with `- ` are required). The calling skill will write this block to `specs/changes/<change-id>/agent-log/contract-reviewer.md`.
-
-```
-## Agent Log
-# Contract Reviewer Log
-- change-id: <id>
-- timestamp: <ISO 8601, e.g. 2026-04-27T14:30:00Z>
-- status: complete | needs-review | blocked
-- artifacts:
-  - <evidence-type>: <concrete pointer>
-  - <evidence-type>: <concrete pointer>
-- next-action: <one line, or "none">
-```
+After completing your task, write or append to
+`specs/changes/<change-id>/agent-log/<your-agent-name>.md`. Required fields,
+field rules, and gate-enforcement behavior are defined once in
+`references/agent-log-protocol.md` — do not duplicate them in this prompt.
 
 ### Required artifacts for this agent
 - `contracts-reviewed`: list of contract file paths
@@ -83,16 +74,6 @@ After completing your task, include an **## Agent Log** section at the end of yo
 - `breaking-changes`: list or "none"
 - `consumers-impacted`: list or "none"
 
-### Rules
-- NEVER omit this log file. `cdd-kit gate` rejects changes whose agent-log
-  is missing the `status:` line or has an invalid status.
-- If you cannot complete the task, set `status: blocked` and write a
-  concrete `next-action` (NOT "investigate further" — write the actual
-  next step a human can act on).
-- Evidence must be concrete: file:line, command name + last-10-line stdout,
-  contract path + section, test name, etc. NEVER write "verified" or "OK"
-  without a pointer.
-
 ## Read scope
 
 - Allowed: `contracts/`, `tests/`, `src/`, `specs/changes/<current-change-id>/`

package/assets/agents/dependency-security-reviewer.md

@@ -64,19 +64,10 @@ approved / changes-required / blocked
 
 ## Machine-Verifiable Evidence
 
-After completing your task, include an **## Agent Log** section at the end of your response with this exact structure (lines starting with `- ` are required). The calling skill will write this block to `specs/changes/<change-id>/agent-log/dependency-security-reviewer.md`.
-
-```
-## Agent Log
-# Dependency Security Reviewer Log
-- change-id: <id>
-- timestamp: <ISO 8601, e.g. 2026-04-27T14:30:00Z>
-- status: complete | needs-review | blocked
-- artifacts:
-  - <evidence-type>: <concrete pointer>
-  - <evidence-type>: <concrete pointer>
-- next-action: <one line, or "none">
-```
+After completing your task, write or append to
+`specs/changes/<change-id>/agent-log/<your-agent-name>.md`. Required fields,
+field rules, and gate-enforcement behavior are defined once in
+`references/agent-log-protocol.md` — do not duplicate them in this prompt.
 
 ### Required artifacts for this agent
 - `packages-reviewed`: list of `<name>@<version>`
@@ -84,12 +75,3 @@ After completing your task, include an **## Agent Log** section at the end of yo
 - `license-issues`: list or "none"
 - `lockfile-changes`: list of files
 
-### Rules
-- NEVER omit this log file. `cdd-kit gate` rejects changes whose agent-log
-  is missing the `status:` line or has an invalid status.
-- If you cannot complete the task, set `status: blocked` and write a
-  concrete `next-action` (NOT "investigate further" — write the actual
-  next step a human can act on).
-- Evidence must be concrete: file:line, command name + last-10-line stdout,
-  contract path + section, test name, etc. NEVER write "verified" or "OK"
-  without a pointer.

package/assets/agents/e2e-resilience-engineer.md

@@ -41,19 +41,10 @@ Record test files, scenarios, fixtures/mocks, commands, screenshots/videos, and
 
 ## Machine-Verifiable Evidence
 
-After completing your task, write or append to `specs/changes/<change-id>/agent-log/<your-agent-name>.md`
-with this exact structure (lines starting with `- ` are required):
-
-```
-# E2E Resilience Engineer Log
-- change-id: <id>
-- timestamp: <ISO 8601, e.g. 2026-04-27T14:30:00Z>
-- status: complete | needs-review | blocked
-- artifacts:
-  - <evidence-type>: <concrete pointer>
-  - <evidence-type>: <concrete pointer>
-- next-action: <one line, or "none">
-```
+After completing your task, write or append to
+`specs/changes/<change-id>/agent-log/<your-agent-name>.md`. Required fields,
+field rules, and gate-enforcement behavior are defined once in
+`references/agent-log-protocol.md` — do not duplicate them in this prompt.
 
 ### Required artifacts for this agent
 - `test-files`: list of paths under `tests/e2e/` or `tests/resilience/`
@@ -61,16 +52,6 @@ with this exact structure (lines starting with `- ` are required):
 - `mutation-checks`: list or "none"
 - `trace-artifacts`: paths or "none"
 
-### Rules
-- NEVER omit this log file. `cdd-kit gate` rejects changes whose agent-log
-  is missing the `status:` line or has an invalid status.
-- If you cannot complete the task, set `status: blocked` and write a
-  concrete `next-action` (NOT "investigate further" — write the actual
-  next step a human can act on).
-- Evidence must be concrete: file:line, command name + last-10-line stdout,
-  contract path + section, test name, etc. NEVER write "verified" or "OK"
-  without a pointer.
-
 ## Read scope
 
 - Allowed: `contracts/`, `tests/`, `src/`, `specs/changes/<current-change-id>/`

package/assets/agents/frontend-engineer.md

@@ -45,19 +45,10 @@ In your agent log, reference file paths and function names — do not paste code
 
 ## Machine-Verifiable Evidence
 
-After completing your task, write or append to `specs/changes/<change-id>/agent-log/<your-agent-name>.md`
-with this exact structure (lines starting with `- ` are required):
-
-```
-# Frontend Engineer Log
-- change-id: <id>
-- timestamp: <ISO 8601, e.g. 2026-04-27T14:30:00Z>
-- status: complete | needs-review | blocked
-- artifacts:
-  - <evidence-type>: <concrete pointer>
-  - <evidence-type>: <concrete pointer>
-- next-action: <one line, or "none">
-```
+After completing your task, write or append to
+`specs/changes/<change-id>/agent-log/<your-agent-name>.md`. Required fields,
+field rules, and gate-enforcement behavior are defined once in
+`references/agent-log-protocol.md` — do not duplicate them in this prompt.
 
 ### Required artifacts for this agent
 - `files-changed`: list of `path/to/file.tsx:line-range`
@@ -65,16 +56,6 @@ with this exact structure (lines starting with `- ` are required):
 - `screenshot-paths`: list of paths under `specs/changes/<id>/screenshots/`
 - `accessibility-audit`: tool name + score or "skipped: reason"
 
-### Rules
-- NEVER omit this log file. `cdd-kit gate` rejects changes whose agent-log
-  is missing the `status:` line or has an invalid status.
-- If you cannot complete the task, set `status: blocked` and write a
-  concrete `next-action` (NOT "investigate further" — write the actual
-  next step a human can act on).
-- Evidence must be concrete: file:line, command name + last-10-line stdout,
-  contract path + section, test name, etc. NEVER write "verified" or "OK"
-  without a pointer.
-
 ## Read scope
 
 - Allowed: `contracts/`, `tests/`, `src/`, `specs/changes/<current-change-id>/`

package/assets/agents/monkey-test-engineer.md

@@ -44,35 +44,16 @@ Use fuzz payloads, Playwright action sequences, property-based tests, and target
 
 ## Machine-Verifiable Evidence
 
-After completing your task, write or append to `specs/changes/<change-id>/agent-log/<your-agent-name>.md`
-with this exact structure (lines starting with `- ` are required):
-
-```
-# Monkey Test Engineer Log
-- change-id: <id>
-- timestamp: <ISO 8601, e.g. 2026-04-27T14:30:00Z>
-- status: complete | needs-review | blocked
-- artifacts:
-  - <evidence-type>: <concrete pointer>
-  - <evidence-type>: <concrete pointer>
-- next-action: <one line, or "none">
-```
+After completing your task, write or append to
+`specs/changes/<change-id>/agent-log/<your-agent-name>.md`. Required fields,
+field rules, and gate-enforcement behavior are defined once in
+`references/agent-log-protocol.md` — do not duplicate them in this prompt.
 
 ### Required artifacts for this agent
 - `test-files`: list of paths under `tests/monkey/`
 - `failure-modes-mapped`: list of `<scenario> → <expected-safe-outcome>`
 - `seeds-recorded`: list of `<test-name>: seed-value` or "deterministic"
 
-### Rules
-- NEVER omit this log file. `cdd-kit gate` rejects changes whose agent-log
-  is missing the `status:` line or has an invalid status.
-- If you cannot complete the task, set `status: blocked` and write a
-  concrete `next-action` (NOT "investigate further" — write the actual
-  next step a human can act on).
-- Evidence must be concrete: file:line, command name + last-10-line stdout,
-  contract path + section, test name, etc. NEVER write "verified" or "OK"
-  without a pointer.
-
 ## Read scope
 
 - Allowed: `contracts/`, `tests/`, `src/`, `specs/changes/<current-change-id>/`

package/assets/agents/qa-reviewer.md

@@ -75,19 +75,10 @@ approved / blocked / approved-with-risk
 
 ## Machine-Verifiable Evidence
 
-After completing your task, include an **## Agent Log** section at the end of your response with this exact structure (lines starting with `- ` are required). The calling skill will write this block to `specs/changes/<change-id>/agent-log/qa-reviewer.md`.
-
-```
-## Agent Log
-# QA Reviewer Log
-- change-id: <id>
-- timestamp: <ISO 8601, e.g. 2026-04-27T14:30:00Z>
-- status: complete | needs-review | blocked
-- artifacts:
-  - <evidence-type>: <concrete pointer>
-  - <evidence-type>: <concrete pointer>
-- next-action: <one line, or "none">
-```
+After completing your task, write or append to
+`specs/changes/<change-id>/agent-log/<your-agent-name>.md`. Required fields,
+field rules, and gate-enforcement behavior are defined once in
+`references/agent-log-protocol.md` — do not duplicate them in this prompt.
 
 ### Required artifacts for this agent
 - `gate-results`: list of `<gate-name>: pass|fail`
@@ -96,16 +87,6 @@ After completing your task, include an **## Agent Log** section at the end of yo
 - `decision`: approved | blocked | approved-with-risk
 - `failure-routing`: list of `<failure-type> → <agent>` or "none"
 
-### Rules
-- NEVER omit this log file. `cdd-kit gate` rejects changes whose agent-log
-  is missing the `status:` line or has an invalid status.
-- If you cannot complete the task, set `status: blocked` and write a
-  concrete `next-action` (NOT "investigate further" — write the actual
-  next step a human can act on).
-- Evidence must be concrete: file:line, command name + last-10-line stdout,
-  contract path + section, test name, etc. NEVER write "verified" or "OK"
-  without a pointer.
-
 ## Read scope
 
 - Allowed: `contracts/`, `tests/`, `src/`, `specs/changes/<current-change-id>/`

package/assets/agents/repo-context-scanner.md

@@ -84,31 +84,13 @@ frontend / backend / fullstack / monorepo / library / tool
 
 ## Machine-Verifiable Evidence
 
-After completing your task, include an **## Agent Log** section at the end of your response with this exact structure (lines starting with `- ` are required). The calling skill will write this block to `specs/changes/<change-id>/agent-log/repo-context-scanner.md`.
-
-```
-## Agent Log
-# Repo Context Scanner Log
-- change-id: <id>
-- timestamp: <ISO 8601, e.g. 2026-04-27T14:30:00Z>
-- status: complete | needs-review | blocked
-- artifacts:
-  - <evidence-type>: <concrete pointer>
-  - <evidence-type>: <concrete pointer>
-- next-action: <one line, or "none">
-```
+After completing your task, write or append to
+`specs/changes/<change-id>/agent-log/<your-agent-name>.md`. Required fields,
+field rules, and gate-enforcement behavior are defined once in
+`references/agent-log-protocol.md` — do not duplicate them in this prompt.
 
 ### Required artifacts for this agent
 - `profile-path`: `project-profile.generated.md`
 - `stack-detected`: from cdd-kit detect-stack
 - `surfaces-flagged`: list of missing standardization surfaces
 
-### Rules
-- NEVER omit this log file. `cdd-kit gate` rejects changes whose agent-log
-  is missing the `status:` line or has an invalid status.
-- If you cannot complete the task, set `status: blocked` and write a
-  concrete `next-action` (NOT "investigate further" — write the actual
-  next step a human can act on).
-- Evidence must be concrete: file:line, command name + last-10-line stdout,
-  contract path + section, test name, etc. NEVER write "verified" or "OK"
-  without a pointer.

package/assets/agents/spec-architect.md

@@ -93,19 +93,10 @@ Target: `design.md` ≤ 150 lines.
 
 ## Machine-Verifiable Evidence
 
-After completing your task, write or append to `specs/changes/<change-id>/agent-log/<your-agent-name>.md`
-with this exact structure (lines starting with `- ` are required):
-
-```
-# Spec Architect Log
-- change-id: <id>
-- timestamp: <ISO 8601, e.g. 2026-04-27T14:30:00Z>
-- status: complete | needs-review | blocked
-- artifacts:
-  - <evidence-type>: <concrete pointer>
-  - <evidence-type>: <concrete pointer>
-- next-action: <one line, or "none">
-```
+After completing your task, write or append to
+`specs/changes/<change-id>/agent-log/<your-agent-name>.md`. Required fields,
+field rules, and gate-enforcement behavior are defined once in
+`references/agent-log-protocol.md` — do not duplicate them in this prompt.
 
 ### Required artifacts for this agent
 - `adr-written`: ADR file path under `docs/adr/` or "no ADR required"
@@ -113,16 +104,6 @@ with this exact structure (lines starting with `- ` are required):
 - `decision-summary`: one-line decision
 - `risks-noted`: count + severity buckets
 
-### Rules
-- NEVER omit this log file. `cdd-kit gate` rejects changes whose agent-log
-  is missing the `status:` line or has an invalid status.
-- If you cannot complete the task, set `status: blocked` and write a
-  concrete `next-action` (NOT "investigate further" — write the actual
-  next step a human can act on).
-- Evidence must be concrete: file:line, command name + last-10-line stdout,
-  contract path + section, test name, etc. NEVER write "verified" or "OK"
-  without a pointer.
-
 ## Read scope
 
 - Allowed: `contracts/`, `tests/`, `src/`, `specs/changes/<current-change-id>/`

package/assets/agents/spec-drift-auditor.md

@@ -52,20 +52,10 @@ By default, do NOT read `specs/changes/` history. Only read historical change re
 
 ## Machine-Verifiable Evidence
 
-Write this block to `specs/audits/<YYYY-MM-DD>-drift-audit.md` (create the file yourself).
-Use this exact structure (lines starting with `- ` are required):
-
-```
-## Agent Log
-# Spec Drift Auditor Log
-- audit-id: <YYYY-MM-DD>-drift
-- timestamp: <ISO 8601, e.g. 2026-04-27T14:30:00Z>
-- status: complete | needs-review | blocked
-- artifacts:
-  - <evidence-type>: <concrete pointer>
-  - <evidence-type>: <concrete pointer>
-- next-action: <one line, or "none">
-```
+After completing your task, write or append to
+`specs/changes/<change-id>/agent-log/<your-agent-name>.md`. Required fields,
+field rules, and gate-enforcement behavior are defined once in
+`references/agent-log-protocol.md` — do not duplicate them in this prompt.
 
 ### Required artifacts for this agent
 - `surfaces-audited`: list (specs/contracts/code/tests/CI/tasks/archive)
@@ -73,11 +63,3 @@ Use this exact structure (lines starting with `- ` are required):
 - `drift-summary-path`: `specs/audits/<YYYY-MM-DD>-drift-audit.md`
 - `next-audit-due`: ISO date
 
-### Rules
-- NEVER omit this audit summary file. The drift-audit cadence (release / weekly / ad-hoc) requires this file as its persistence record; missing `status:` voids the audit.
-- If you cannot complete the task, set `status: blocked` and write a
-  concrete `next-action` (NOT "investigate further" — write the actual
-  next step a human can act on).
-- Evidence must be concrete: file:line, command name + last-10-line stdout,
-  contract path + section, test name, etc. NEVER write "verified" or "OK"
-  without a pointer.

package/assets/agents/stress-soak-engineer.md

@@ -68,19 +68,10 @@ Use realistic load profiles rather than arbitrary request loops.
 
 ## Machine-Verifiable Evidence
 
-After completing your task, write or append to `specs/changes/<change-id>/agent-log/<your-agent-name>.md`
-with this exact structure (lines starting with `- ` are required):
-
-```
-# Stress Soak Engineer Log
-- change-id: <id>
-- timestamp: <ISO 8601, e.g. 2026-04-27T14:30:00Z>
-- status: complete | needs-review | blocked
-- artifacts:
-  - <evidence-type>: <concrete pointer>
-  - <evidence-type>: <concrete pointer>
-- next-action: <one line, or "none">
-```
+After completing your task, write or append to
+`specs/changes/<change-id>/agent-log/<your-agent-name>.md`. Required fields,
+field rules, and gate-enforcement behavior are defined once in
+`references/agent-log-protocol.md` — do not duplicate them in this prompt.
 
 ### Required artifacts for this agent
 - `runner-config-path`: e.g. `tests/stress/<scenario>.js`
@@ -88,16 +79,6 @@ with this exact structure (lines starting with `- ` are required):
 - `pass-criteria-cited`: SLO references (must include p95 / error-rate / leak-signal numbers)
 - `artifacts-location`: path
 
-### Rules
-- NEVER omit this log file. `cdd-kit gate` rejects changes whose agent-log
-  is missing the `status:` line or has an invalid status.
-- If you cannot complete the task, set `status: blocked` and write a
-  concrete `next-action` (NOT "investigate further" — write the actual
-  next step a human can act on).
-- Evidence must be concrete: file:line, command name + last-10-line stdout,
-  contract path + section, test name, etc. NEVER write "verified" or "OK"
-  without a pointer.
-
 ## Read scope
 
 - Allowed: `contracts/`, `tests/`, `src/`, `specs/changes/<current-change-id>/`