deploy-fix（部署修复）

Deploy Fix Skill

Diagnose a build, CI, or deployment failure, choose the safest repair path, and fix it — classify complexity first, then either fix directly or track progress through structured phases.

<request_routing>

Positive triggers

• A build command fails with a concrete error such as Module not found, type errors, or compilation failures.
• A CI pipeline step fails with a specific error in logs (lint, test, build, or deploy stage).
• A deployment fails with a concrete error such as function timeout, missing env vars, or platform-specific build errors.
• A specific folder or workspace within a monorepo fails to build.

Out-of-scope

• Runtime bugs in application code with a reproduction path. Route to bug-fix.
• Security audits, exploit review, or trust-boundary analysis. Route to security-review.
• New feature work, refactors, or speculative cleanup not tied to a concrete failure.
• General performance optimization without a failing build or deploy.

Boundary cases

• If the user asks for root-cause analysis only, stay in diagnosis mode and do not edit.
• If a CI failure is caused by a single runtime bug (e.g., a failing test from a code defect), this skill owns the CI-level fix; hand off the underlying code bug to bug-fix if the root cause is application logic.
• If the failure spans build + deployment + runtime, own the build/deploy layer and hand off runtime to bug-fix.

</request_routing>

<argument_validation>

If ARGUMENT is missing, ask immediately:

What build/CI/deploy failure should be fixed?
- Error message or failing log output
- Build command or CI step that fails
- Full repo build or specific folder/workspace?
- CI provider (GitHub Actions, Vercel, etc.) if applicable
- Regression or new? (Did it work before? When did it start failing?)
- Consistent or intermittent?
- Recent changes, suspect commits, or environment details
- Relevant config files (package.json, tsconfig, CI config, vercel.json, etc.)

</argument_validation>

<mandatory_reasoning>

Mandatory Sequential Thinking

Always run sequential-thinking before implementation. Depth scales with complexity:

• Simple (3 thoughts): Identify failing step -> determine fix -> verify approach
• Medium (5 thoughts): Classify -> reproduce locally -> hypothesize -> compare options -> recommend
• Complex (7+ thoughts): Classify -> reproduce -> analyze dependency chain -> check CI config -> hypothesize multiple causes -> compare options -> assess cross-cutting impact -> recommend

Recommended sequence:

1. Complexity classification
2. Failure reproduction and log analysis
3. Root-cause hypotheses
4. Option comparison
5. Final recommendation

Before any edit, collect root-cause evidence from build logs, CI output, or deployment logs and reduce the problem to the narrowest failing boundary you can verify.

</mandatory_reasoning>

<complexity_classification>

Complexity Classification

Classify immediately after sequential-thinking:

Announce the classification:

Complexity: [simple/complex] -- [one-line reason]

When uncertain, classify as complex. It is cheaper to track than to lose investigation progress.

</complexity_classification>

<flow_tracking>

Flow Tracking (Complex Path Only)

When classified as complex, initialize the flow:

mkdir -p .hypercore/deploy-fix

Write .hypercore/deploy-fix/flow.json and update it as each phase progresses. See references/flow-schema.md for the full schema.

Phase progression

Resume support

If .hypercore/deploy-fix/flow.json already exists, read it first and continue from the last incomplete phase (in_progress or pending). Do not restart completed phases.

</flow_tracking>

<execution_modes>

Use one of these branches explicitly:

• Diagnose-only: reproduce failure, isolate the failing step, summarize evidence, and stop before code edits.
• Fix-now (simple path): If the user explicitly asks for a direct fix and one path is clearly the safest, say which path you are taking and implement without a second confirmation round. No flow tracking.
• Option-first (complex path): present 2-3 repair options with flow tracking and wait for user selection.
• Handoff: route runtime application bugs to bug-fix and security review requests to security-review.

</execution_modes>

<investigation_strategy>

Investigation Strategy

Check these areas in order of likelihood:

1. Build logs: read the exact error output, identify the first failure point
2. Dependency issues: lockfile conflicts (package-lock.json/pnpm-lock.yaml integrity), missing packages, version mismatches, peer dependency warnings, hoisting problems
3. Config files: tsconfig.json, package.json, next.config.*, vercel.json/vercel.ts, CI workflow files, bundler config
4. Environment: Node.js version, env vars, platform-specific differences (local vs CI vs deploy target), memory limits (OOM during build)
5. Build cache: stale or corrupted build cache (.next/, .turbo/, node_modules/.cache/), --force rebuild to isolate
6. Cross-workspace: monorepo dependency graph, build order, shared package version drift
7. Recent changes: git log and git diff to find suspect commits that may have introduced the failure

</investigation_strategy>

Simple Path (Fix-now)

Complex Path (Option-first)

<option_presentation>

Use this format (complex path):

## Deploy Failure Analysis

**Root cause**: ...
**Failure scope**: [repo-wide / workspace / CI step / deploy target]
**Complexity**: complex

### Option 1: ... (Recommended)
- **Pros**:
- **Cons**:
- **Risk**:
- **Files**:

### Option 2: ...
- **Pros**:
- **Cons**:
- **Risk**:
- **Files**:

### Option 3: ... (Temporary)
- **Pros**:
- **Cons**:
- **Risk**:
- **Files**:

Recommendation: Option N (reason ...)
Which option should I apply? (1/2/3)

</option_presentation>

<implementation_rules>

• Do not modify code before user option selection unless in the explicit Fix-now branch.
• Avoid speculative edits; use evidence from build/CI/deploy logs only.
• Keep scope limited to the failing build/CI/deploy path and its direct dependencies.
• Always run targeted validation for the changed path: rebuild the failing target, re-run the failing CI step, or re-trigger the failing deploy.
• Report the commands run, the key result lines, and the touched files in the final report.
• If validation cannot run locally (e.g., CI-only environment), say why and what remains unverified.

Reporting

After execution, report:

## Done

**Failure**: [original error / failing step]
**Root cause**: [what was wrong]
**Fix applied**: [which option or approach]
**Changes**: [list of changed files]
**Validation**: [what was verified and result]

For complex path: also update .hypercore/deploy-fix/flow.json status to completed.

</implementation_rules>

Execution checklist:

• ARGUMENT validated
• sequential-thinking completed (depth matches complexity)
• Complexity classified (simple/complex)
• Flow JSON created and maintained (complex path only)
• Root-cause evidence collected from logs/config
• 2-3 options presented (complex path) or fix path announced (simple path)
• User choice confirmed (complex path)
• Build/CI/deploy validation executed
• Outcome + touched files reported
• Flow JSON finalized with completed status (complex path only)

Forbidden:

• Speculative fix without build/CI/deploy log evidence
• Immediate implementation without options (complex path)
• Implementation without explicit user choice (complex path)
• Completion claim without running the failing build/CI/deploy command
• Skipping flow JSON updates in complex path