contract-driven-delivery 2.0.0 → 2.0.1

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/CHANGELOG.md CHANGED
@@ -1,5 +1,17 @@
1
1
  # Changelog
2
2
 
3
+ ## [2.0.1] - 2026-04-30
4
+
5
+ ### Fixed
6
+
7
+ - Clarified the agent ownership model in the public docs so read-only reviewers
8
+ and write-capable implementation agents have explicit, non-conflicting file
9
+ ownership rules.
10
+ - Aligned bundled prompts so read-only agents emit an `Agent Log` YAML block
11
+ for main Claude to persist, while write-capable agents continue writing their
12
+ own artifacts and `agent-log/*.yml` files.
13
+ - Synchronized package version metadata for the post-`2.0.0` publish path.
14
+
3
15
  ## [2.0.0] - 2026-04-30
4
16
 
5
17
  ### BREAKING: structured YAML for tasks and agent-log
package/README.md CHANGED
@@ -99,6 +99,19 @@ or
99
99
  8. `cdd-kit gate <change-id>` runs automatically to confirm all artifacts are complete
100
100
  9. Claude reports a summary and the suggested git commit
101
101
 
102
+ ### Agent Ownership Model
103
+
104
+ CDD uses two agent classes on purpose:
105
+
106
+ - `change-classifier`, `contract-reviewer`, `qa-reviewer`, `visual-reviewer`, `dependency-security-reviewer`, `ui-ux-reviewer`, `repo-context-scanner`, and `spec-drift-auditor` are read-only. They return analysis, verdicts, or an `Agent Log` YAML block; main Claude writes the corresponding files.
107
+ - `backend-engineer`, `frontend-engineer`, `e2e-resilience-engineer`, `monkey-test-engineer`, `stress-soak-engineer`, `ci-cd-gatekeeper`, `test-strategist`, and `spec-architect` are write-capable. They write their own implementation artifacts and their own `agent-log/*.yml`.
108
+
109
+ This split is deliberate:
110
+
111
+ - Review and audit agents stay read-only so they do not silently change the thing they are supposed to assess.
112
+ - Implementation and planning agents write directly so large artifacts and code edits do not have to be relayed back through the main orchestrator, which reduces token waste and preserves clearer ownership.
113
+ - `tasks.yml` remains owned by main Claude so task state changes stay centralized even when multiple agents contribute files.
114
+
102
115
  **You stay in control by:**
103
116
  - Reviewing the `change-classification.md` before implementation starts
104
117
  - Checking the `test-plan.md` to confirm the right test families are planned
@@ -18,12 +18,12 @@ Read `specs/changes/<change-id>/context-manifest.md` before using file-reading o
18
18
  - Read only paths allowed by the manifest or approved expansions.
19
19
  - Do not use broad repository search unless the manifest authorizes it.
20
20
  - If more context is needed, stop and write a Context Expansion Request in the manifest.
21
- - Record every file read through tools in the relevant `agent-log/*.md` under `- files-read:`.
21
+ - Record every file read through tools in the relevant `agent-log/*.yml` under `files-read:`.
22
22
 
23
- Required `agent-log/*.md` format:
23
+ Required `agent-log/*.yml` format:
24
24
 
25
- ```md
26
- - files-read:
25
+ ```yaml
26
+ files-read:
27
27
  - contracts/api/api-contract.md
28
28
  - src/server/routes/users.ts
29
29
  ```
@@ -225,7 +225,8 @@ Note: `archive.md` is created during change close-out, not at classification tim
225
225
 
226
226
  ## Machine-Verifiable Evidence
227
227
 
228
- After completing your task, write or append to
228
+ After completing your task, end your response with an `Agent Log` YAML block
229
+ for main Claude to write to
229
230
  `specs/changes/<change-id>/agent-log/<your-agent-name>.yml`. Required fields,
230
231
  field rules, and gate-enforcement behavior are defined once in
231
232
  `references/agent-log-protocol.md` — do not duplicate them in this prompt.
@@ -63,7 +63,8 @@ approved / changes-required
63
63
 
64
64
  ## Machine-Verifiable Evidence
65
65
 
66
- After completing your task, write or append to
66
+ After completing your task, end your response with an `Agent Log` YAML block
67
+ for main Claude to write to
67
68
  `specs/changes/<change-id>/agent-log/<your-agent-name>.yml`. Required fields,
68
69
  field rules, and gate-enforcement behavior are defined once in
69
70
  `references/agent-log-protocol.md` — do not duplicate them in this prompt.
@@ -64,7 +64,8 @@ approved / changes-required / blocked
64
64
 
65
65
  ## Machine-Verifiable Evidence
66
66
 
67
- After completing your task, write or append to
67
+ After completing your task, end your response with an `Agent Log` YAML block
68
+ for main Claude to write to
68
69
  `specs/changes/<change-id>/agent-log/<your-agent-name>.yml`. Required fields,
69
70
  field rules, and gate-enforcement behavior are defined once in
70
71
  `references/agent-log-protocol.md` — do not duplicate them in this prompt.
@@ -74,4 +75,3 @@ field rules, and gate-enforcement behavior are defined once in
74
75
  - `cve-findings`: count + severity buckets
75
76
  - `license-issues`: list or "none"
76
77
  - `lockfile-changes`: list of files
77
-
@@ -75,7 +75,8 @@ approved / blocked / approved-with-risk
75
75
 
76
76
  ## Machine-Verifiable Evidence
77
77
 
78
- After completing your task, write or append to
78
+ After completing your task, end your response with an `Agent Log` YAML block
79
+ for main Claude to write to
79
80
  `specs/changes/<change-id>/agent-log/<your-agent-name>.yml`. Required fields,
80
81
  field rules, and gate-enforcement behavior are defined once in
81
82
  `references/agent-log-protocol.md` — do not duplicate them in this prompt.
@@ -84,7 +84,8 @@ frontend / backend / fullstack / monorepo / library / tool
84
84
 
85
85
  ## Machine-Verifiable Evidence
86
86
 
87
- After completing your task, write or append to
87
+ After completing your task, end your response with an `Agent Log` YAML block
88
+ for main Claude to write to
88
89
  `specs/changes/<change-id>/agent-log/<your-agent-name>.yml`. Required fields,
89
90
  field rules, and gate-enforcement behavior are defined once in
90
91
  `references/agent-log-protocol.md` — do not duplicate them in this prompt.
@@ -93,4 +94,3 @@ field rules, and gate-enforcement behavior are defined once in
93
94
  - `profile-path`: `project-profile.generated.md`
94
95
  - `stack-detected`: from cdd-kit detect-stack
95
96
  - `surfaces-flagged`: list of missing standardization surfaces
96
-
@@ -52,7 +52,8 @@ By default, do NOT read `specs/changes/` history. Only read historical change re
52
52
 
53
53
  ## Machine-Verifiable Evidence
54
54
 
55
- After completing your task, write or append to
55
+ After completing your task, end your response with an `Agent Log` YAML block
56
+ for main Claude to write to
56
57
  `specs/changes/<change-id>/agent-log/<your-agent-name>.yml`. Required fields,
57
58
  field rules, and gate-enforcement behavior are defined once in
58
59
  `references/agent-log-protocol.md` — do not duplicate them in this prompt.
@@ -62,4 +63,3 @@ field rules, and gate-enforcement behavior are defined once in
62
63
  - `drift-items`: count + severity
63
64
  - `drift-summary-path`: `specs/audits/<YYYY-MM-DD>-drift-audit.md`
64
65
  - `next-audit-due`: ISO date
65
-
@@ -51,7 +51,8 @@ approved / changes-required
51
51
 
52
52
  ## Machine-Verifiable Evidence
53
53
 
54
- After completing your task, write or append to
54
+ After completing your task, end your response with an `Agent Log` YAML block
55
+ for main Claude to write to
55
56
  `specs/changes/<change-id>/agent-log/<your-agent-name>.yml`. Required fields,
56
57
  field rules, and gate-enforcement behavior are defined once in
57
58
  `references/agent-log-protocol.md` — do not duplicate them in this prompt.
@@ -61,4 +62,3 @@ field rules, and gate-enforcement behavior are defined once in
61
62
  - `state-coverage`: list of `<screen>: empty/loading/error/success` matrix
62
63
  - `copy-issues`: count + severity
63
64
  - `accessibility-findings`: count + severity
64
-
@@ -53,7 +53,8 @@ approved / changes-required
53
53
 
54
54
  ## Machine-Verifiable Evidence
55
55
 
56
- After completing your task, write or append to
56
+ After completing your task, end your response with an `Agent Log` YAML block
57
+ for main Claude to write to
57
58
  `specs/changes/<change-id>/agent-log/<your-agent-name>.yml`. Required fields,
58
59
  field rules, and gate-enforcement behavior are defined once in
59
60
  `references/agent-log-protocol.md` — do not duplicate them in this prompt.
@@ -63,4 +64,3 @@ field rules, and gate-enforcement behavior are defined once in
63
64
  - `diff-percentage`: per-screen
64
65
  - `state-coverage`: matrix
65
66
  - `tokens-violated`: list of CSS contract violations or "none"
66
-
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "contract-driven-delivery",
3
- "version": "2.0.0",
3
+ "version": "2.0.1",
4
4
  "description": "Contract-driven delivery kit for AI coding agents with deterministic context indexes, manifest-backed read-scope governance, and orchestrated contracts-first delivery.",
5
5
  "keywords": [
6
6
  "contract-driven",