cflx-workflow（CFLX 工作流程）

Conflux Workflow Executor

Execute Conflux workflow operations autonomously. Called by orchestration system, not for direct human use.

CRITICAL: This skill CANNOT ask questions to users. All decisions must be made autonomously based on available context.

Operation Modes

This skill supports three operations, determined by the orchestrator's invocation:

1. apply - Implement an approved change
2. accept - Verify implementation against specs
3. archive - Finalize a deployed change

Operation Selection

The orchestrator specifies the operation. Parse the invocation to determine:

• If change ID with "apply" or "implement" context → Execute Apply
• If "accept" or "review" context → Execute Accept
• If "archive" context → Execute Archive

Operation 1: Apply (Implementation)

Purpose: Implement an approved change autonomously with task tracking.

CRITICAL CONSTRAINTS:

• NO QUESTIONS - Make autonomous decisions based on available context
• NO DEFERRAL - Do not defer tasks based on difficulty or complexity
• IMMEDIATE UPDATES - Update tasks.md after EVERY completed task
• COMPLETE ALL TRUTHFULLY - A task may be marked [x] only when the corresponding repository change and required verification actually exist
• ESCALATE BLOCKERS - If implementation is impossible, record an Implementation Blocker for acceptance review
• NO CHECKLIST-ONLY COMPLETION - Do not mark implementation tasks complete based only on proposal/spec/tasks edits when the task requires code, tests, or runtime wiring

Execution Steps

1. Read Proposal python3 "$SKILL_ROOT/scripts/cflx.py" show <change-id>

- Read openspec/changes/<id>/proposal.md - Read openspec/changes/<id>/design.md (if exists) - Read openspec/changes/<id>/tasks.md

1. Work Through Tasks Sequentially

- Start with first uncompleted task - Implement the change - Run verification (build/test/lint) - Mark task as [x] in tasks.md immediately after the implementation and verification evidence exist - Proceed to next task

1. Handle Ambiguity Autonomously

- Use existing code patterns as reference - Make reasonable assumptions - Document decisions in code comments - Prefer simpler solutions

1. Update Progress Continuously

- Update tasks.md after each task - Never batch updates - Keep progress visible

1. Verify Completion

- Ensure all tasks are [x] or in Future Work - Run final validation - Confirm integration points

Truthful Completion Rules

Before changing any task to [x], verify all applicable conditions below are true:

1. The repository contains the required implementation artifact for that task.

- Code task -> matching src/, app, config, or script diff exists. - Test task -> matching tests/ diff exists. - Wiring/integration task -> real entrypoint/call-site/config hookup exists. - Spec-only task -> it is explicitly documentation/spec work rather than implementation work.

1. The artifact is reachable from the intended flow when the task claims runtime integration.
2. The relevant verification command has been run successfully, or concrete blocker evidence has been recorded.
3. The task description still matches reality. If the task is too broad or ambiguous, refine it before completion.

Never mark a task complete based only on any of the following:

• openspec/ files were updated
• tasks.md was normalized
• a proposal was archived or merged
• code was discussed but no runtime/test artifact was added
• a stub placeholder was added where a real execution path was required

Task Management

Move to Future Work ONLY if:

1. Requires human decision-making or judgment
2. Requires external system access outside repository
3. Requires long-wait verification (>1 day)
4. Already marked with '(future work)'

Do NOT move to Future Work:

• Difficult or complex tasks (agent must attempt)
• Tests (unit/integration/e2e)
• Linting/formatting
• Documentation updates
• Any automatable task

Checkbox Rules

Active sections: Must have checkboxes - [] or - [x]

Excluded sections (Future Work, Out of Scope, Notes): Must NOT have checkboxes

## Implementation Tasks
- [x] Completed task
- [ ] Pending task

## Future Work
- Manual verification required
- External deployment needed

Mock-First Policy

• Mock external dependencies when possible
• Do not block on missing API keys/credentials
• Implement stub/fixture for external services
• Only truly non-mockable dependencies go to Future Work

Implementation Blocker Escalation

If apply determines the change is currently impossible to implement (for example: spec contradiction, non-mockable external limitation, or policy constraint), do not loop blindly.

1. Add a new section to openspec/changes/<change-id>/tasks.md: ## Implementation Blocker #<n> - category: <spec_contradiction|external_non_mockable|policy_constraint|other> - summary: <one-line human-facing blocker summary> - evidence: - <file/path:line or concrete command output> - impact: <what cannot be completed> - unblock_actions: - <specific follow-up action 1> - <specific follow-up action 2> - owner: <team_or_role> - decision_due: <YYYY-MM-DD>
2. The blocker section is human-facing and MUST NOT use checkboxes.
3. Output a machine-readable marker at the end of apply output: IMPLEMENTATION_BLOCKER: category: <...> tasks_section: "Implementation Blocker #<n>" human_action_required: see openspec/changes/<change-id>/tasks.md#implementation-blocker-<n>
4. Keep evidence concrete and actionable so acceptance can judge whether loop stop is warranted.

Apply Completion Criteria

• All tasks marked [x] or moved to Future Work (without checkboxes)
• Code compiles/builds successfully
• Tests pass
• Lint passes
• Integration points verified
• Any task that claims implementation, runtime behavior, or entrypoint wiring has corresponding non-OpenSpec evidence in the repo
• Changes that are spec-only MUST leave implementation tasks unchecked or blocked; they must not be represented as completed implementation

For detailed guidance, read references/cflx-apply.md.

Operation 2: Accept (Acceptance Review)

Purpose: Verify implementation meets specifications with automated checks.

CRITICAL: Output exactly ONE verdict at the end.

Required Checks

1. Git Working Tree Clean git status --porcelain

- Must output empty (no uncommitted changes) - If dirty, output FAIL with all changed files

1. Task Completion

- All tasks marked [x] or in Future Work section - No checkboxes in excluded sections - Reject any task marked [x] without corresponding repo evidence

1. Spec Matching

- Implementation matches specification in specs/ - All scenarios are satisfied

1. Integration Check

- Feature is executed in real flow - Called from CLI/TUI/API as specified

1. Dead Code Check

- All implemented code is invoked - No orphan functions/classes

1. Regression Check

- Existing features still work - No unintended side effects

1. Evidence Citation

- Cite file path + function/method for integration - Provide concrete verification evidence

1. Checklist Truthfulness Check

- FAIL if tasks.md claims completion but the corresponding code/tests/entrypoints do not exist - FAIL if a change was archived/spec-promoted while implementation tasks were marked complete without repository evidence - FAIL if the only evidence for an implementation task is openspec/ edits

Output Format

Output exactly ONE of these at the end:

PASS:

ACCEPTANCE: PASS

FAIL:

ACCEPTANCE: FAIL

FINDINGS:
1. [file:line] Description of issue
2. [file:line] Description of issue
...

Then update tasks.md with:

## Acceptance #N Failure Follow-up

- [ ] Fix issue 1
- [ ] Fix issue 2

CONTINUE (only if verification incomplete):

ACCEPTANCE: CONTINUE

BLOCKED (when blocker escalation is valid):

ACCEPTANCE: BLOCKED

BLOCKER:
- category: <...>
- reason: <short rationale>
- evidence: <file/path:line or command evidence>

Recommended:
- summary: <one-line human-facing blocker summary>
- unblock_actions:
  - <specific follow-up action 1>
  - <specific follow-up action 2>

Accept Rules

• Each finding must include concrete evidence (file path, function, line)
• Each finding must be actionable by AI agent
• Missing secrets MUST NOT cause CONTINUE if mocking is possible
• Dirty working tree is always FAIL
• ACCEPTANCE: BLOCKED is allowed only when a valid Implementation Blocker #<n> exists with concrete evidence and unblock actions
• If blocker data is weak, speculative, or fixable within repo scope, return FAIL instead of BLOCKED

For detailed guidance, read references/cflx-accept.md.

Operation 3: Archive

Purpose: Archive deployed change and update canonical specs.

Execution Steps

1. Identify Change ID

- From orchestrator invocation - Or from context (must be unambiguous)

1. Validate Change Status python3 "$SKILL_ROOT/scripts/cflx.py" list python3 "$SKILL_ROOT/scripts/cflx.py" show <id>

- Ensure change exists - Ensure not already archived - Ensure ready for archive

1. Run Archive python3 "$SKILL_ROOT/scripts/cflx.py" archive <id> --yes

- Use --skip-specs only for tooling-only changes

1. Verify Results python3 "$SKILL_ROOT/scripts/cflx.py" validate --strict

- Confirm moved to changes/archive/ - Confirm specs updated

Archive Completion Criteria

• Change moved to openspec/changes/archive/<id>/
• Canonical specs updated (unless --skip-specs)
• Validation passes with --strict

For detailed guidance, read references/cflx-archive.md.

Built-in Tools

# List changes
python3 "$SKILL_ROOT/scripts/cflx.py" list

# List specs
python3 "$SKILL_ROOT/scripts/cflx.py" list --specs

# Show change details
python3 "$SKILL_ROOT/scripts/cflx.py" show <id>

# Show JSON output
python3 "$SKILL_ROOT/scripts/cflx.py" show <id> --json

# Show deltas only
python3 "$SKILL_ROOT/scripts/cflx.py" show <id> --json --deltas-only

# Validate change
python3 "$SKILL_ROOT/scripts/cflx.py" validate <id> --strict

# Validate all
python3 "$SKILL_ROOT/scripts/cflx.py" validate --strict

# Archive change
python3 "$SKILL_ROOT/scripts/cflx.py" archive <id> --yes

# Archive without spec updates
python3 "$SKILL_ROOT/scripts/cflx.py" archive <id> --yes --skip-specs

Autonomous Decision Framework

When facing ambiguous situations, follow this priority:

1. Existing patterns - Follow patterns in the codebase
2. Specification - Refer to spec deltas and scenarios
3. Simplicity - Choose simpler implementation
4. Documentation - Document decision in code comments

Never:

• Ask user for clarification
• Stop and wait for input
• Leave tasks incomplete due to uncertainty

Task Format Requirements

Valid:

- [ ] Task description
- [x] Completed task
1. [ ] Numbered task

Invalid (must fix):

## N. Task              → - [ ] N. Task
- Task                 → - [ ] Task
1. Task                → 1. [ ] Task

If 0/0 tasks detected, fix format first.

Error Handling

Validation Failure

1. Parse error messages
2. Fix identified issues
3. Re-run validation
4. Repeat until passing

Build/Test Failure

1. Analyze error output
2. Fix code issues
3. Re-run verification
4. Update tasks on success

Incomplete Information

1. Make reasonable assumption
2. Implement based on assumption
3. Document assumption in code
4. Continue with next task

Reference Files

Detailed operation guides:

• references/cflx-apply.md - Apply operation details
• references/cflx-accept.md - Accept operation details
• references/cflx-archive.md - Archive operation details

Summary

REMEMBER: This skill operates autonomously. Never ask questions. Make decisions based on available context.