agentic-devtools 0.2.111__tar.gz → 0.2.112__tar.gz

{agentic_devtools-0.2.111 → agentic_devtools-0.2.112}/.github/copilot-instructions.md

@@ -126,6 +126,20 @@ Why this matters:
 - **Atomic changes** — each PR represents a single logical change
 - **Easier reverts** — one commit = one revert if something goes wrong
 
+### ⚠️ CRITICAL: Auto-Merge Label (`ai-auto-merge-allowed`)
+
+**Before kicking off any AI agent session that will result in a PR being created, you MUST ask the user whether the `ai-auto-merge-allowed` label should be added to the PR.**
+
+Rules for AI agents:
+
+- **NEVER add `ai-auto-merge-allowed` to a PR unless you were explicitly instructed to do so** in the session instructions.
+- When starting an agent session (e.g. `/agdt.address-copilot-review`, `agdt-initiate-*-workflow`, `agdt.pr-merge-manager`, or any other workflow that will result in a PR), **ask the user first**:
+  > "Should the `ai-auto-merge-allowed` label be added to the PR? (This enables auto-merge via the AI PR loop once all checks pass and Copilot approves.)"
+- If the user says yes, **include an explicit instruction in the agent's session prompt**: `"Add the 'ai-auto-merge-allowed' label to the PR after creating it."`
+- If the user says no or does not answer, **do not add the label**.
+- Repo workflows may auto-apply this label only when explicitly enabled (for example `SPECKIT_AUTO_MERGE_ALLOWED_LABEL == 'true'`); if not enabled, agents must treat it as not approved.
+- Do not infer intent from context — only add the label when explicitly instructed.
+
 ### ⚠️ CRITICAL: Always Use agdt Commands
 
 **AI agents MUST prefer `agdt-*` CLI commands over raw equivalents wherever support exists.**

{agentic_devtools-0.2.111 → agentic_devtools-0.2.112}/.github/workflows/README.md

@@ -19,7 +19,7 @@ This directory contains GitHub Actions workflows for the agentic-devtools projec
 2. **Trigger Guards**: Prevents redundant runs — if triggered by Copilot review but checks are still pending,
    polls every 30 seconds for up to 10 minutes (skips on timeout);
    if a Copilot review on the head commit has `CHANGES_REQUESTED` state, blocks merge until resolved
-3. **Exclusion Labels**: `ai-pr-loop-ignore` skips entirely; `do-not-auto-merge` prevents merge but allows fixes/approval
+3. **Exclusion Labels**: `ai-pr-loop-ignore` skips entirely; missing `ai-auto-merge-allowed` prevents merge but allows fixes/approval
 4. **Repair dispatch**: Uses failed PR checks and Copilot review state to decide when to request fixes
 5. **Amend & Push**: Amends fixes into the last commit with `--force-with-lease`
 6. **Copilot Review Handling**: Detects outstanding Copilot review comments and blocks merge until resolved

{agentic_devtools-0.2.111 → agentic_devtools-0.2.112}/.github/workflows/speckit-issue-trigger.yml

@@ -38,9 +38,49 @@ jobs:
             catch (error) { core.warning(`Failed to add speckit:processing label to issue #${issueNumber}: ${error.message}`); }
       - uses: actions/setup-python@v5
         with: { python-version: '3.12' }
-      - run: python -m pip install --upgrade pip && python -m pip install . --no-deps
-      - env: { GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}, COPILOT_GITHUB_TOKEN: ${{ secrets.COPILOT_GITHUB_TOKEN }}, SPECKIT_TRIGGER_LABEL: ${{ vars.SPECKIT_TRIGGER_LABEL || 'speckit' }} }
+      - run: |
+          python -m pip install --upgrade pip
+          python -m pip install github-copilot-sdk
+          python -m pip install . --no-deps
+      - env:
+          GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+          COPILOT_GITHUB_TOKEN: ${{ secrets.COPILOT_GITHUB_TOKEN }}
+          SPECKIT_TRIGGER_LABEL: ${{ vars.SPECKIT_TRIGGER_LABEL || 'speckit' }}
         run: agdt-speckit-trigger
+      - name: Add ai-auto-merge-allowed label to Phase 1 PR
+        if: success() && vars.SPECKIT_AUTO_MERGE_ALLOWED_LABEL == 'true'
+        env:
+          ISSUE_NUMBER: ${{ github.event.issue.number || inputs.issue_number }}
+        uses: actions/github-script@v7
+        with:
+          script: |
+            const issueNumber = Number(process.env.ISSUE_NUMBER);
+            if (!Number.isInteger(issueNumber) || issueNumber <= 0) {
+              core.warning('Could not determine issue number for speckit phase-1 PR lookup');
+              return;
+            }
+
+            try {
+              const query = `repo:${context.repo.owner}/${context.repo.repo} is:pr is:open head:speckit/${issueNumber}/phase-1-specify`;
+              const search = await github.rest.search.issuesAndPullRequests({
+                q: query,
+                per_page: 1,
+              });
+              const speckit1Pr = search.data.items?.[0];
+              if (!speckit1Pr) {
+                core.warning(`Could not find speckit phase-1 PR for issue #${issueNumber} to label`);
+                return;
+              }
+              await github.rest.issues.addLabels({
+                owner: context.repo.owner,
+                repo: context.repo.repo,
+                issue_number: speckit1Pr.number,
+                labels: ['ai-auto-merge-allowed'],
+              });
+              core.info(`Added ai-auto-merge-allowed to PR #${speckit1Pr.number}`);
+            } catch (e) {
+              core.warning(`Could not add ai-auto-merge-allowed label: ${e.message}`);
+            }
 
       - name: Handle Failure (Comment + Label)
         if: failure()

{agentic_devtools-0.2.111 → agentic_devtools-0.2.112}/.github/workflows/speckit-phase-progression.yml

@@ -599,6 +599,35 @@ jobs:
               "${LINT_ARGS[@]}"
           fi
 
+      - name: Add ai-auto-merge-allowed label (non-clarify phases)
+        if: >-
+          steps.extract.outcome == 'success' &&
+          steps.extract.outputs.next_phase != '0' &&
+          steps.extract.outputs.next_phase != '6' &&
+          vars.SPECKIT_AUTO_MERGE_ALLOWED_LABEL == 'true' &&
+          steps.idempotency.outputs.skipped != 'true' &&
+          steps.create-pr.outputs.pr_number != '' &&
+          steps.extract.outputs.next_phase_name != 'clarify'
+        uses: actions/github-script@v7
+        with:
+          script: |
+            const prNumber = parseInt('${{ steps.create-pr.outputs.pr_number }}', 10);
+            if (!prNumber) {
+              core.warning('No PR number available — skipping ai-auto-merge-allowed label');
+              return;
+            }
+            try {
+              await github.rest.issues.addLabels({
+                owner: context.repo.owner,
+                repo: context.repo.repo,
+                issue_number: prNumber,
+                labels: ['ai-auto-merge-allowed'],
+              });
+              core.info(`Added ai-auto-merge-allowed to PR #${prNumber}`);
+            } catch (e) {
+              core.warning(`Could not add ai-auto-merge-allowed label: ${e.message}`);
+            }
+
       - name: Auto-Merge (if configured)
         if: steps.extract.outcome == 'success' && steps.extract.outputs.next_phase != '6' && steps.extract.outputs.next_phase != '0' && steps.idempotency.outputs.skipped != 'true' && steps.create-pr.outputs.pr_number != '' && steps.generate.outputs.gate_result != 'fail'
         env:

{agentic_devtools-0.2.111 → agentic_devtools-0.2.112}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: agentic-devtools
-Version: 0.2.111
+Version: 0.2.112
 Summary: Agentic devtools integrate Jira, DevOps & more
 Author: ayaiayorg
 License-Expression: MIT

{agentic_devtools-0.2.111 → agentic_devtools-0.2.112}/agentic_devtools/_version.py

@@ -18,7 +18,7 @@ version_tuple: tuple[int | str, ...]
 commit_id: str | None
 __commit_id__: str | None
 
-__version__ = version = '0.2.111'
-__version_tuple__ = version_tuple = (0, 2, 111)
+__version__ = version = '0.2.112'
+__version_tuple__ = version_tuple = (0, 2, 112)
 
 __commit_id__ = commit_id = None

{agentic_devtools-0.2.111 → agentic_devtools-0.2.112}/agentic_devtools/cli/ci/ado_provider.py

@@ -113,7 +113,7 @@ class AzureDevOpsProvider(CIPlatformProvider):
         """Not implemented for ADO stub."""
         raise NotImplementedError("AzureDevOpsProvider.find_comment() not yet implemented")
 
-    def approve_pr(self, pr_number: int, head_sha: str, body: str) -> None:
+    def approve_pr(self, pr_number: int, head_sha: str, body: str) -> bool:
         """Not implemented for ADO stub."""
         raise NotImplementedError("AzureDevOpsProvider.approve_pr() not yet implemented")
 

{agentic_devtools-0.2.111 → agentic_devtools-0.2.112}/agentic_devtools/cli/ci/github_provider.py

@@ -500,12 +500,15 @@ class GitHubActionsProvider(CIPlatformProvider):
         return None
 
     @retry_with_backoff()
-    def approve_pr(self, pr_number: int, head_sha: str, body: str) -> None:
+    def approve_pr(self, pr_number: int, head_sha: str, body: str) -> bool:
         """Approve a pull request.
 
         Uses ``AGDT_PR_APPROVER_PAT`` when set so that the approval comes from
         a separate identity (GitHub prevents approving your own PR).  When the
         variable is unset or empty the approval is skipped with a warning.
+
+        Returns:
+            ``True`` when approval was submitted, ``False`` when skipped.
         """
         approver_token = os.environ.get("AGDT_PR_APPROVER_PAT", "").strip()
         if not approver_token:
@@ -514,7 +517,7 @@ class GitHubActionsProvider(CIPlatformProvider):
                 "Cannot approve PR without a dedicated approver token. "
                 "See repository documentation for setup instructions."
             )
-            return
+            return False
 
         try:
             _gh_api(
@@ -523,6 +526,7 @@ class GitHubActionsProvider(CIPlatformProvider):
                 body={"commit_id": head_sha, "event": "APPROVE", "body": body},
                 token=approver_token,
             )
+            return True
         except RuntimeError as exc:
             stderr = str(exc)
             if "401" in stderr or "Bad credentials" in stderr.lower():
@@ -531,7 +535,7 @@ class GitHubActionsProvider(CIPlatformProvider):
                     "The token may be expired or invalid. "
                     "Skipping PR approval. Rotate the secret and retry."
                 )
-                return
+                return False
             raise
 
     @retry_with_backoff()

{agentic_devtools-0.2.111 → agentic_devtools-0.2.112}/agentic_devtools/cli/ci/guards.py

@@ -28,7 +28,7 @@ DOCKER_PATTERNS = (re.compile(r"^\.dockerignore$"), re.compile(r"^Dockerfile\..*
 
 # Labels that affect PR processing
 LABEL_SKIP_ENTIRELY = "ai-pr-loop-ignore"
-LABEL_NO_AUTO_MERGE = "do-not-auto-merge"
+LABEL_AUTO_MERGE_ALLOWED = "ai-auto-merge-allowed"
 
 # Deduplication marker format
 DEDUP_MARKER_PREFIX = "<!-- repair-dispatch:"
@@ -176,11 +176,12 @@ def check_exclusion_labels(labels: list[str]) -> tuple[bool, str | None]:
         Tuple of (should_skip, flag_name).
         - should_skip=True, flag_name=None: skip entirely (ai-pr-loop-ignore)
         - should_skip=False, flag_name="do_not_merge": process but don't merge
+          (ai-auto-merge-allowed label missing)
         - should_skip=False, flag_name=None: no exclusion
     """
     if LABEL_SKIP_ENTIRELY in labels:
         return (True, None)
-    if LABEL_NO_AUTO_MERGE in labels:
+    if LABEL_AUTO_MERGE_ALLOWED not in labels:
         return (False, "do_not_merge")
     return (False, None)
 

{agentic_devtools-0.2.111 → agentic_devtools-0.2.112}/agentic_devtools/cli/ci/orchestrator.py

@@ -22,7 +22,6 @@ from typing import Any
 from agentic_devtools.cli.ci.guards import (
     DEDUP_MARKER_PATTERN,
     DEDUP_MARKER_PREFIX,
-    LABEL_NO_AUTO_MERGE,
     LABEL_SKIP_ENTIRELY,
     PRIVILEGED_PREFIXES,
     check_cycle_limit,
@@ -164,18 +163,16 @@ def _apply_comment_triggered_post_repair_guards(
         summary["exit_code"] = EXIT_GUARD_BLOCKED
         return EXIT_GUARD_BLOCKED
 
-    should_skip, flag = check_exclusion_labels(pr_meta.labels)
-    if should_skip or flag == "do_not_merge":
-        blocked_by = "exclusion_label" if should_skip else "do_not_merge"
-        label = LABEL_SKIP_ENTIRELY if should_skip else LABEL_NO_AUTO_MERGE
+    should_skip, _ = check_exclusion_labels(pr_meta.labels)
+    if should_skip:
         logger.info(
             "PR #%d comment-triggered post-repair squash blocked by label guard (%s)",
             pr_number,
-            label,
+            LABEL_SKIP_ENTIRELY,
         )
-        summary["guards"] = {"blocked_by": blocked_by, "label": label}
+        summary["guards"] = {"blocked_by": "exclusion_label", "label": LABEL_SKIP_ENTIRELY}
         summary["decision"] = "blocked"
-        summary["reason"] = blocked_by
+        summary["reason"] = "exclusion_label"
         summary["exit_code"] = EXIT_GUARD_BLOCKED
         return EXIT_GUARD_BLOCKED
 
@@ -967,6 +964,45 @@ def run_ai_pr_loop(
                 _emit_decision_summary(summary)
                 return EXIT_SUCCESS
 
+        # CI passed + clean Copilot review on HEAD -> approve so the review gate
+        # re-runs on the pull_request_review event and can turn green.
+        _copilot_review_ci = latest_by_user.get(_COPILOT_KEY)
+        has_clean_copilot_review_on_head = (
+            _copilot_review_ci is not None
+            and (not _copilot_review_ci.commit_sha or _copilot_review_ci.commit_sha == current_head_sha)
+            and _copilot_review_ci.state in ("APPROVED", "COMMENTED")
+            and not copilot_actionable_review
+        )
+        approval_skipped_in_ci_completion = False
+        if has_clean_copilot_review_on_head:
+            logger.info(
+                "PR #%d CI passed with clean Copilot review on HEAD — approving to trigger review gate re-run",
+                pr_number,
+            )
+            try:
+                approved = provider.approve_pr(pr_number, pr_meta.head_sha, "Auto-approved by AI PR loop")
+            except Exception as exc:
+                logger.error("Failed to auto-approve PR #%d: %s", pr_number, exc)
+                summary["decision"] = "error"
+                summary["reason"] = "approval_failed"
+                summary["error"] = str(exc)
+                summary["exit_code"] = EXIT_MERGE_BLOCKED
+                _log_endgroup()
+                _emit_decision_summary(summary)
+                return EXIT_MERGE_BLOCKED
+            if approved:
+                summary["auto_approved"] = True
+                summary["decision"] = "approved_awaiting_review_gate"
+                summary["exit_code"] = EXIT_SUCCESS
+                _log_endgroup()
+                _emit_decision_summary(summary)
+                return EXIT_SUCCESS
+            logger.info(
+                "PR #%d clean Copilot review found, but auto-approval was skipped; awaiting manual approval",
+                pr_number,
+            )
+            approval_skipped_in_ci_completion = True
+
         # No actionable review and no failures — ensure a Copilot review is
         # requested so the gate can pass on the current HEAD.
         if pr_meta.is_draft:
@@ -982,6 +1018,15 @@ def run_ai_pr_loop(
             _emit_decision_summary(summary)
             return EXIT_SUCCESS
         summary["repair"] = {"needed": False}
+        if approval_skipped_in_ci_completion:
+            # Clean Copilot review exists but approval was skipped (e.g., missing approver token).
+            # Don't request another review — it already exists and won't re-trigger the gate.
+            summary["decision"] = "awaiting_copilot_review_after_ci"
+            summary["reason"] = "approval_skipped"
+            summary["exit_code"] = EXIT_SUCCESS
+            _log_endgroup()
+            _emit_decision_summary(summary)
+            return EXIT_SUCCESS
         logger.info("PR #%d CI passed, no Copilot review on HEAD — requesting review", pr_number)
         request_skip_reason = _request_copilot_review_if_needed(
             provider,
@@ -1077,9 +1122,9 @@ def run_ai_pr_loop(
         return EXIT_SUCCESS
 
     if do_not_merge:
-        logger.info("PR #%d ready but do-not-auto-merge label present — skipping merge", pr_number)
+        logger.info("PR #%d ready but 'ai-auto-merge-allowed' label not present — skipping merge", pr_number)
         summary["decision"] = "skip_merge"
-        summary["reason"] = "do_not_auto_merge_label"
+        summary["reason"] = "missing_auto_merge_allowed_label"
         summary["exit_code"] = EXIT_SUCCESS
         _log_endgroup()
         _emit_decision_summary(summary)
@@ -1172,7 +1217,7 @@ def run_ai_pr_loop(
     if not has_approval:
         logger.info("PR #%d has no approval — auto-approving", pr_number)
         try:
-            provider.approve_pr(pr_number, pr_meta.head_sha, "Auto-approved by AI PR loop")
+            approved = provider.approve_pr(pr_number, pr_meta.head_sha, "Auto-approved by AI PR loop")
         except Exception as exc:
             logger.error("Failed to auto-approve PR #%d: %s", pr_number, exc)
             summary["decision"] = "error"
@@ -1182,6 +1227,18 @@ def run_ai_pr_loop(
             _log_endgroup()
             _emit_decision_summary(summary)
             return EXIT_MERGE_BLOCKED
+        if not approved:
+            logger.warning(
+                "PR #%d auto-approval skipped; blocking merge until approval can be submitted",
+                pr_number,
+            )
+            summary["auto_approved"] = False
+            summary["decision"] = "blocked"
+            summary["reason"] = "approval_skipped"
+            summary["exit_code"] = EXIT_MERGE_BLOCKED
+            _log_endgroup()
+            _emit_decision_summary(summary)
+            return EXIT_MERGE_BLOCKED
         summary["auto_approved"] = True
     else:
         summary["auto_approved"] = False

{agentic_devtools-0.2.111 → agentic_devtools-0.2.112}/agentic_devtools/cli/ci/provider.py

@@ -108,13 +108,16 @@ class CIPlatformProvider(ABC):
         """
 
     @abstractmethod
-    def approve_pr(self, pr_number: int, head_sha: str, body: str) -> None:
+    def approve_pr(self, pr_number: int, head_sha: str, body: str) -> bool:
         """Approve a pull request.
 
         Args:
             pr_number: Pull request number.
             head_sha: Expected HEAD SHA (for safety check).
             body: Approval comment body.
+
+        Returns:
+            ``True`` when approval was posted, ``False`` when intentionally skipped.
         """
 
     @abstractmethod

agentic_devtools-0.2.112/agentic_devtools/prompts/ci/ready-no-merge-comment.md

@@ -0,0 +1,12 @@
+<!-- markdownlint-disable MD041 -->
+✅ **AI PR Loop — Ready (No Auto-Merge)**
+
+This PR is ready to merge but does not have the `ai-auto-merge-allowed` label applied.
+
+**PR**: #{{pr_number}}
+**Branch**: `{{head_branch}}`
+**SHA**: `{{head_sha}}`
+**CI Status**: All checks passing
+**Reviews**: Copilot review complete
+
+Add the `ai-auto-merge-allowed` label to allow automated merge, or merge manually.

agentic_devtools-0.2.112/specs/1430-langchain-based-work-issue/analysis-report.md

@@ -0,0 +1,119 @@
+# Cross-Artifact Consistency & Quality Analysis Report
+
+## Findings Table
+
+| ID | Category | Severity | Location(s) | Summary | Recommendation |
+|----|----------|----------|-------------|---------|----------------|
+| F-01 | B | MEDIUM | Spec: NFR-001 | "same order of magnitude" and "no more than 2x slower" — the 2x threshold is defined but "excluding external API latency" makes measurement ambiguous (how to isolate API latency in practice?) | Add a concrete measurement methodology (e.g., "measured by subtracting HTTP round-trip time from wall-clock time" or "measured using mocked tool layer") |
+| F-02 | B | LOW | Spec: NFR-002 | "same conventions as the existing workflow (progress messages, step announcements, error formatting)" — no specific format contract defined | Consider adding example output format or referencing existing output format documentation |
+| F-03 | C | MEDIUM | Spec: FR-003, Plan Phase 3 | `tools.jira.fetch_issue_context()` and `tools.jira.add_comment()` are referenced but no contract/signature is defined for these functions — it's unclear if they exist today or must be created | Clarify whether these tool-layer functions already exist or are new deliverables; if new, add tasks for their creation |
+| F-04 | C | MEDIUM | Spec: FR-003, Plan Phase 3 | `tools.azure_devops.create_pull_request()` is referenced as a synchronous function but the existing codebase uses background-task patterns for PR creation — unclear if this function exists or needs an adapter | Document whether this is an existing function or requires a new synchronous adapter |
+| F-05 | F | MEDIUM | Plan Phase 3 vs Spec | Plan references `setup_node` consuming "explicit pre-flight results from runner startup" but spec User Story 1 acceptance scenario 1 doesn't mention a separate setup node — spec flows directly from "initiate through completion" listing setup as a graph node | Minor inconsistency; both align on setup existing as a node but spec could explicitly mention the runner-to-node handoff pattern |
+| F-06 | F | LOW | Tasks T055 vs Plan Phase 3 | T055 says "Write tests for the pre-flight node behavior" but doesn't name the node function (`setup_node`) — inconsistent with T068 which names it explicitly | Rename T055 to reference `setup_node()` explicitly for clarity |
+| F-07 | C | MEDIUM | Spec: US3 AC2 | `tools.git.save_work()` and individual functions like `stage_changes()`, `create_commit()`, `amend_commit()`, `push()`, `force_push()` — unclear which API surface the commit_node actually calls (the aggregate or the individual functions) | Spec should pick one canonical call pattern for `commit_node`; plan chose `save_work()` adapter — update spec AC to match |
+| F-08 | G | HIGH | Tasks T030, T105, T106 | T030 "Write tests for `run_langchain_workflow()` fresh invocation" overlaps with T106 "Write happy-path success test for `run_langchain_workflow()` fresh invocation" — same function, same fresh-invocation scenario | Consolidate T030 and T106 into a single test task covering both error and happy-path fresh invocation, or clarify that T030 is purely negative-path |
+| F-09 | G | HIGH | Tasks T033, T091 | T033 "Write tests for resume with no existing checkpoint — exit code 1" and T091 "Write tests for `--resume` with no interrupted workflow — exit code 1" — same scenario (FR-012), same expected behavior | Consolidate into one task or clearly differentiate scope (runner-level vs integration-level) |
+| F-10 | G | HIGH | Tasks T034, T107 | T034 "Write tests for resume from `planning_gate` — `Command(resume=True)` invocation" and T107 "Write happy-path success test for planning gate resume" — same gate, same resume mechanism | Differentiate: T034 should focus on the `Command` invocation mechanics, T107 on end-to-end flow continuation |
+| F-11 | G | HIGH | Tasks T036, T108 | T036 "Write tests for resume from `implementation_gate` with valid `--resume-data`" and T108 "Write happy-path success test for implementation gate resume" — same gate, same valid data scenario | Consolidate or explicitly scope T036 to runner internals and T108 to CLI-through-runner integration |
+| F-12 | G | HIGH | Tasks T035, T090 | T035 "Write tests for resume from `implementation_gate` without `--resume-data` — exit code 1" and T090 "Write tests for process restart resume from implementation gate checkpoint with `--resume-data`" — partial overlap on implementation gate resume testing | Clarify: T035 is negative-path (missing data), T090 is positive-path (with data after restart) — add explicit differentiation in descriptions |
+| F-13 | A | LOW | Spec: US4 AC1 vs US1 AC2 | US4 AC1 ("workflow is paused at the planning gate, process restarted, resumed") largely restates US1 AC2 with the addition of "process restart" — near-duplicate acceptance criteria | Keep US4 AC1 but add explicit "process was killed and restarted" precondition to differentiate from in-session resume |
+| F-14 | F | LOW | Plan Phase 5 vs Tasks | Plan estimates "~15-20 new test files" while tasks enumerate ~67 test tasks; these are different units and may be compatible if multiple tasks are grouped per file | Clarify the grouping assumption explicitly (e.g., "~67 test tasks implemented across ~15-20 test files") to avoid a misleading mismatch signal |
+| F-15 | D | LOW | Spec | No explicit "Out of Scope" section — constitution/quality-gate best practice for feature specs | Add an "Out of Scope" section explicitly listing what this feature does NOT cover (e.g., multi-model orchestration, parallel node execution) |
+
+<!-- markdownlint-disable MD013 MD033 -->
+
+### Category G Structured Findings
+
+[
+  {
+    "id": "F-08",
+    "overlap_type": "overlapping",
+    "severity": "HIGH",
+    "task_ids": ["T030", "T105", "T106"],
+    "dimensions": ["description"],
+    "rationale": "T030 and T106 both target 'run_langchain_workflow() fresh invocation' scenarios. T105 also covers fresh-invocation routing to the runner. The overlap classification is based on highly similar descriptions and scenario intent, not file-path overlap."
+  },
+  {
+    "id": "F-09",
+    "overlap_type": "overlapping",
+    "severity": "HIGH",
+    "task_ids": ["T033", "T091"],
+    "dimensions": ["description"],
+    "rationale": "Both tasks test the exact same FR-012 scenario: --resume with no existing checkpoint yields exit code 1 with descriptive error. T033 is in Phase 3 (runner tests) and T091 in Phase 6 (checkpoint persistence tests) but the test logic and assertion are identical."
+  },
+  {
+    "id": "F-10",
+    "overlap_type": "overlapping",
+    "severity": "HIGH",
+    "task_ids": ["T034", "T107"],
+    "dimensions": ["description"],
+    "rationale": "Both test resume from planning_gate with Command(resume=True). T034 focuses on invocation mechanics, while T107 focuses on happy-path continuation. The overlap is description-level and gate-level; no explicit shared file path is asserted."
+  },
+  {
+    "id": "F-11",
+    "overlap_type": "overlapping",
+    "severity": "HIGH",
+    "task_ids": ["T036", "T108"],
+    "dimensions": ["description"],
+    "rationale": "Both test resume from implementation_gate with valid --resume-data producing Command(resume=<dict>). They share the same gate and valid-data scenario, and differ only implicitly between 'mechanics' and 'happy-path success'."
+  },
+  {
+    "id": "F-12",
+    "overlap_type": "overlapping",
+    "severity": "HIGH",
+    "task_ids": ["T035", "T090"],
+    "dimensions": ["description"],
+    "rationale": "T035 tests implementation_gate resume without --resume-data (negative). T090 tests implementation_gate resume WITH --resume-data after process restart (positive). They overlap on gate and resume context, but do not claim explicit shared file-path overlap."
+  }
+]
+
+<!-- markdownlint-enable MD013 MD033 -->
+
+## Coverage Summary Table
+
+| Requirement Key | Has Task? | Task IDs | Notes |
+|----------------|-----------|----------|-------|
+| FR-001 | ✅ | T016-T029, T030-T050, T105-T112 | Fully covered |
+| FR-002 | ✅ | T051-T053, T085, T113-T114 | Fully covered |
+| FR-003 | ✅ | T014-T015, T054-T078, T079-T085 | Fully covered |
+| FR-004 | ✅ | T034, T039, T046, T050, T057, T060, T080, T088-T092, T104, T107 | Fully covered |
+| FR-005 | ✅ | T004-T007, T043, T088-T092 | Fully covered |
+| FR-006 | ✅ | T086-T087, T110 | Fully covered |
+| FR-007 | ✅ | T008-T011, T054-T078, T066, T111 | Fully covered |
+| FR-008 | ✅ | T062, T075, T081, T084, T112 | Fully covered |
+| FR-009 | ✅ | T012-T013, T031-T032, T040 | Fully covered |
+| FR-010 | ✅ | T053, T103, T114 | Fully covered |
+| FR-011 | ✅ | T018, T026, T109 | Fully covered |
+| FR-012 | ✅ | T033, T037, T045, T091, T109 | Fully covered |
+| FR-013 | ✅ | T019-T021, T027, T035-T036, T047, T090, T108 | Fully covered |
+| NFR-001 | ✅ | T095 | Single integration test |
+| NFR-002 | ✅ | T049 | Implementation task, no dedicated test |
+| NFR-003 | ✅ | T004-T007 | Fully covered |
+| NFR-004 | ✅ | T101 | Full suite validation |
+| NFR-005 | ✅ | T103 | Verification task |
+
+## Metrics
+
+| Metric | Value |
+|--------|-------|
+| Total Requirements | 18 (13 FR + 5 NFR) |
+| Total Tasks | 115 (T001–T103, T104–T115) |
+| Coverage % | 100% |
+| Ambiguity Count | 2 (F-01, F-02) |
+| Requirement Duplication Count (Category A) | 1 (F-13) |
+| Critical Issues Count | 0 |
+| Task Deduplication Finding Count | 5 (F-08 through F-12) |
+| Task Deduplication by Type | duplicate: 0 / overlapping: 5 / conflicting: 0 |
+| Multi-Task Group Count | 0 |
+
+---
+*Generated by Copilot SDK (claude-opus-4.6)*
+
+## Next Actions
+
+1. Resolve the HIGH-severity task overlaps first, especially F-08 through F-12, by consolidating duplicate test tasks or making their scopes explicitly distinct.
+2. Clarify ambiguous tool contracts and API surfaces called out in F-03, F-04, and F-07 before implementation begins, so plan/tasks align with actual code interfaces.
+3. Tighten spec language for measurement and output-format expectations noted in F-01 and F-02 to reduce implementation ambiguity and test gaps.
+4. Apply minor consistency cleanups such as the naming alignment noted in F-05 and F-06 to improve traceability across spec, plan, and tasks.
+
+Would you like me to suggest concrete remediation edits for the spec, plan, and tasks files to address the findings above?

{agentic_devtools-0.2.111 → agentic_devtools-0.2.112}/tests/unit/cli/ci/github_provider/test_approve_pr.py

@@ -24,7 +24,7 @@ class TestApprovePR:
         provider = GitHubActionsProvider(repo="owner/repo")
         result = provider.approve_pr(42, "abc123", "LGTM!")
 
-        assert result is None
+        assert result is True
         kwargs = mock_run_safe.call_args[1]
         body = json.loads(kwargs["input"])
         assert body["event"] == "APPROVE"
@@ -38,8 +38,9 @@ class TestApprovePR:
     def test_approve_pr_skips_when_pat_empty(self, mock_run_safe) -> None:
         """Approval is skipped when AGDT_PR_APPROVER_PAT is empty."""
         provider = GitHubActionsProvider(repo="owner/repo")
-        provider.approve_pr(42, "abc123", "LGTM!")
+        result = provider.approve_pr(42, "abc123", "LGTM!")
 
+        assert result is False
         mock_run_safe.assert_not_called()
 
     @patch.dict("os.environ", {"AGDT_PR_APPROVER_PAT": "not_present"})
@@ -51,8 +52,9 @@ class TestApprovePR:
         # Remove the key after patch.dict sets it, simulating unset env var
         del os.environ["AGDT_PR_APPROVER_PAT"]
         provider = GitHubActionsProvider(repo="owner/repo")
-        provider.approve_pr(42, "abc123", "LGTM!")
+        result = provider.approve_pr(42, "abc123", "LGTM!")
 
+        assert result is False
         mock_run_safe.assert_not_called()
 
     @patch.dict("os.environ", {"AGDT_PR_APPROVER_PAT": "ghp_expired_token"})
@@ -68,8 +70,8 @@ class TestApprovePR:
         mock_run_safe.return_value = _Result()
 
         provider = GitHubActionsProvider(repo="owner/repo")
-        # Should not raise
-        provider.approve_pr(42, "abc123", "LGTM!")
+        result = provider.approve_pr(42, "abc123", "LGTM!")
+        assert result is False
 
     @patch.dict("os.environ", {"AGDT_PR_APPROVER_PAT": "ghp_token"})
     @patch("agentic_devtools.cli.ci.github_provider.run_safe")

{agentic_devtools-0.2.111 → agentic_devtools-0.2.112}/tests/unit/cli/ci/guards/test_check_exclusion_labels.py

@@ -11,29 +11,29 @@ class TestCheckExclusionLabels:
         assert should_skip is True
         assert flag is None
 
-    def test_no_merge_label_sets_flag(self) -> None:
-        should_skip, flag = check_exclusion_labels(["do-not-auto-merge"])
+    def test_missing_auto_merge_allowed_label_sets_flag(self) -> None:
+        should_skip, flag = check_exclusion_labels(["bug"])
         assert should_skip is False
         assert flag == "do_not_merge"
 
-    def test_no_exclusion_labels(self) -> None:
-        should_skip, flag = check_exclusion_labels(["bug", "enhancement"])
+    def test_auto_merge_allowed_label_clears_do_not_merge_flag(self) -> None:
+        should_skip, flag = check_exclusion_labels(["bug", "enhancement", "ai-auto-merge-allowed"])
         assert should_skip is False
         assert flag is None
 
     def test_empty_labels(self) -> None:
         should_skip, flag = check_exclusion_labels([])
         assert should_skip is False
-        assert flag is None
+        assert flag == "do_not_merge"
 
     def test_ignore_takes_priority(self) -> None:
         """When both labels are present, ignore takes priority."""
-        should_skip, flag = check_exclusion_labels(["ai-pr-loop-ignore", "do-not-auto-merge"])
+        should_skip, flag = check_exclusion_labels(["ai-pr-loop-ignore", "ai-auto-merge-allowed"])
         assert should_skip is True
         assert flag is None
 
     def test_case_sensitive(self) -> None:
-        """Labels are case-sensitive."""
-        should_skip, flag = check_exclusion_labels(["AI-PR-LOOP-IGNORE"])
+        """Labels are case-sensitive — wrong case of ignore label does not trigger skip."""
+        should_skip, flag = check_exclusion_labels(["AI-PR-LOOP-IGNORE", "ai-auto-merge-allowed"])
         assert should_skip is False
         assert flag is None