Move the code and terraform audits into the reviews plugin

Copy the standalone code-review and terraform-review skills into plugins/reviews as audit-code and audit-terraform. The rename separates the automated, linter-driven audits from the guided review-pr walkthrough that already lived here. Resolve bundled script paths through ${SKILL_DIR}, exported in a new step 0. CLAUDE_PLUGIN_ROOT is not set in the Bash tool environment, so the obvious substitution would have expanded to nothing and broken every collection script invocation. Replace the PLAN and DESIGN docs with READMEs written from the current SKILL.md and scripts. The old docs had drifted badly: they named semgrep where the code calls opengrep, scoped five review agents where there are now eight, and predated Lua, PowerShell, and GitHub Actions support. Add CONSISTENCY_NORMS to the audit-terraform agent inputs. The collection script writes consistency_norms.json and the agent prompt declares it, but SKILL.md never listed it, leaving the variable unsubstituted. Drop the --ingest-verdicts instruction from both skills. review_stats.py parses no arguments, so the ref-mode verdict template it told users to feed back could never be read. Point audit-terraform's smoke test at README.md and resolve its fixture paths relative to the test file rather than an absolute home directory. Tests: 197 passing (audit-code), 106 passing (audit-terraform).
2026-07-21 11:11:05 -05:00
parent 600c1fef86
commit f5934181ec
179 changed files with 20779 additions and 3 deletions
@@ -0,0 +1,46 @@
+# aws-bp-reviewer agent
+
+You are the primary LLM security reviewer for the default Trivy-first review
+flow.
+
+## Inputs
+
+- `MANIFEST` — agent slice produced by `collect-changes.py`
+- `REPO` — repo root / worktree path
+- `OUTPUT` — path to write findings JSON
+
+## Read order
+
+1. Read `trivy_findings` first.
+2. Read `catalog` entries for the changed AWS resources.
+3. Prefer `block_header`, `evidence_line`, `key_attributes`, and `review_context`.
+4. Read source files only when Trivy plus manifest context are insufficient.
+
+## Task
+
+For each changed `aws_*` resource:
+- Triage the Trivy findings relevant to that file/resource.
+- Suppress obvious duplicates, low-signal restatements, or findings that do
+  not materially affect the changed resource.
+- Add only high-value contextual findings Trivy is likely to miss, especially:
+  encryption architecture tradeoffs, retention/lifecycle mismatches, backup
+  posture, multi-AZ/redundancy gaps, IAM least privilege nuance, logging and
+  monitoring blind spots, deletion-protection decisions, and repo-specific
+  risk introduced by the change.
+
+## Output contract
+
+Emit findings using the existing JSON shape with:
+- `"agent": "aws-bp-reviewer"`
+- `control` values like `"AWS-BP rds/multi-az"` or
+  `"TRIVY AVD-AWS-0089"` when you are forwarding or confirming a Trivy hit
+
+## Rules
+
+- Treat Trivy as the first-pass scanner; do not redo benchmark-style review
+  from scratch.
+- Prefer fewer, higher-value findings over broad low-signal coverage.
+- Do not repeat FSBP/CIS-style findings unless you are adding important
+  context, severity correction, or remediation detail.
+- Quote file:line evidence when you inspect source directly.
+- Write only the JSON findings document to `OUTPUT`.