tanviet12/vbsec / skills/antigravity/vbs-scan-security
vbs-scan-security
FlaggedUse when scanning code for security vulnerabilities. Use when user says "scan security", "kiểm tra bảo mật", "security audit", "review security", or invokes `/vbs-scan-security`. For large scans (>20 main-language files OR >30 total OR >14 days) processes chunks sequentially. Outputs bilingual reports (vi/en). Optional `--auto-fix` (agentic patch + verify loop) and `--sca` (live CVE lookup via OSV.dev).
references/i18n/en.md
View on GitHub# i18n — English
Key→text mapping for report output when `lang=en`. SKILL.md and workflows must read this file and substitute keys into output. Do NOT hardcode English strings in logic.
## Header
| Key | Text |
|---|---|
| `report_title` | vbsec Security Scan Report |
| `header_scope` | Scope |
| `header_files` | Files |
| `header_primary_lang` | Primary language |
| `header_mode` | Mode |
| `header_date` | Scan date |
| `header_lang` | Report language |
| `header_using_generic` | using generic rules |
| `header_using_specialized` | using specialized rules |
| `header_no_specialized` | no specialized rules available |
## Scope labels
| Key | Text |
|---|---|
| `scope_uncommitted` | Uncommitted changes |
| `scope_staged` | Staged files only |
| `scope_commit_within` | Commits within last {days} days |
| `scope_commit_id` | Commit {sha} |
| `scope_pr` | Pull Request #{number} |
| `scope_all` | Entire repo |
## Mode
| Key | Text |
|---|---|
| `mode_small` | SMALL (inline scan) |
| `mode_large` | LARGE (sub-agent delegation) |
## Verdict
| Key | Text |
|---|---|
| `verdict_label` | VERDICT |
| `verdict_pass` | PASS |
| `verdict_warn` | PASS WITH WARNINGS |
| `verdict_fail` | FAIL |
| `verdict_pass_desc` | No critical issues found. Safe to deploy. |
| `verdict_warn_desc` | HIGH-severity issues found. Address before production. |
| `verdict_fail_desc` | CRITICAL issues found. DO NOT deploy until fixed. |
## Severity
| Key | Text |
|---|---|
| `severity_critical` | CRITICAL |
| `severity_high` | HIGH |
| `severity_medium` | MEDIUM |
| `severity_low` | LOW |
| `severity_critical_short` | Critical |
| `severity_high_short` | High |
| `severity_medium_short` | Medium |
| `severity_low_short` | Low |
## Section headers
| Key | Text |
|---|---|
| `section_critical` | CRITICAL (blocks deploy) |
| `section_high` | HIGH (must fix) |
| `section_medium` | MEDIUM |
| `section_low` | LOW |
| `section_passed` | PASSED CHECKS |
| `section_warnings` | WARNINGS |
| `section_summary` | Summary |
| `section_recommendations` | Recommendations |
## Table headers
| Key | Text |
|---|---|
| `col_file_line` | File:Line |
| `col_rule` | Rule |
| `col_issue` | Issue |
| `col_fix` | Fix |
| `col_severity` | Severity |
| `col_recommendation` | Recommendation |
| `col_count` | Count |
| `col_status` | Status |
## Counts & summary
| Key | Text |
|---|---|
| `count_files_reviewed` | Files reviewed |
| `count_rules_applied` | Rules applied |
| `count_findings` | Total findings |
| `count_critical` | Critical issues |
| `count_high` | High issues |
| `count_medium` | Medium issues |
| `count_low` | Low issues |
| `no_findings` | No issues found |
| `all_passed` | All checks passed |
## Common phrases (for fix recommendations — rule files can template these)
| Key | Text |
|---|---|
| `phrase_use_parameterized` | Use parameterized query / prepared statement |
| `phrase_never_concat` | Never concatenate untrusted input into queries |
| `phrase_use_env_var` | Move to environment variable; do not commit secrets |
| `phrase_rotate_now` | Rotate the key immediately (if it was pushed to Git) |
| `phrase_add_gitignore` | Add `.env` to `.gitignore` |
| `phrase_validate_input` | Validate and sanitize input before use |
| `phrase_use_bcrypt` | Use bcrypt/argon2/scrypt for password hashing |
| `phrase_use_csrf_token` | Add CSRF token for state-changing endpoints |
| `phrase_specific_origin` | Set a specific origin instead of `*` |
| `phrase_use_transaction` | Wrap in a transaction with row-level lock to prevent races |
| `phrase_check_authz` | Add backend authorization check; never trust the frontend |
| `phrase_disable_debug_prod` | Disable `DEBUG=true` in production |
| `phrase_update_dep` | Update the dependency to a CVE-patched version |
| `phrase_use_safe_yaml` | Use `yaml.safe_load` (Python) or equivalent |
| `phrase_no_shell_true` | Do not use `shell=True` with user input |
| `phrase_whitelist_redirect` | Whitelist redirect destination domains |
## System messages
| Key | Text |
|---|---|
| `msg_no_git` | Not a git repository. Please `cd` into a git repo first. |
| `msg_scope_needs_git` | This scope requires a git repository. Run the skill with no scope (defaults to scanning the entire folder), or init git first. Tip: `git init && git add .` |
| `msg_no_git_note` | ⚠️ Not a git repository — scanned all files in the folder. Did not respect `.gitignore` (none applicable). |
| `msg_no_files` | No files in scan scope. |
| `msg_routing_small` | SMALL mode: inline scan |
| `msg_routing_large` | LARGE mode: chunking into {n} groups, spawning sub-agents |
| `msg_scanning` | Scanning... |
| `msg_done` | Scan complete |
| `msg_invalid_lang` | Invalid `lang` parameter. Use `vi` or `en`. Default: `vi`. |
## Footer
| Key | Text |
|---|---|
| `footer_generated_by` | Generated by |
| `footer_repo` | https://github.com/tanviet12/vbsec |
| `footer_disclaimer` | This report is a reference — not a substitute for professional security audit. |
| `footer_next_steps` | Next steps |
| `footer_next_steps_text` | Fix CRITICAL issues first. Re-scan to confirm. |
## Verbose finding blocks (v0.3+)
| Key | Text |
|---|---|
| `header_short_description` | Short description |
| `header_why_dangerous` | Why is this dangerous? |
| `header_attack_scenario` | How does an attacker exploit this? |
| `header_current_code` | Current code (UNSAFE) |
| `header_safe_code` | Safe code (replace with) |
| `header_impact` | Impact |
| `header_fix_code` | Fix |
| `header_read_more` | Read more |
| `header_overview_table` | Overview (details below) |
| `phrase_input_untrusted` | input from user — UNTRUSTED |
| `phrase_concat_to_sink` | concatenated directly into |
## Persistence (v0.3+)
| Key | Text |
|---|---|
| `msg_report_saved` | Report saved to |
| `msg_gitignore_warning_title` | Recommendation |
| `msg_gitignore_warning_text` | The `vbsec-reports/` directory is not listed in `.gitignore`. To avoid committing scan reports to Git, add `vbsec-reports/` to `.gitignore`. |
| `msg_reports_dir_created` | Created `vbsec-reports/` directory for persisted reports |
## Hardening notes
| Key | Text |
|---|---|
| `header_hardening_title` | Hardening notes (not vulnerabilities) |
| `msg_hardening_intro` | The items below are not exploitable; they are optional defense-in-depth suggestions and do not affect the verdict. |
| `header_version` | vbsec version |
| `header_context` | Project rules |
| `header_coverage_title` | Scan coverage |
| `msg_coverage_intro` | What was read in full, what was only pattern-searched, which rules were not loaded. Issues in parts not read in full may be missed. |
| `label_other_cwe` | Outside the rule catalog (still counts toward the verdict) |
## Auto-fix (`--auto-fix`, v0.7+)
| Key | Text |
|---|---|
| `header_autofix_title` | Auto-fix results |
| `msg_autofix_needs_git` | ⚠️ `--auto-fix` requires a git repository (patches are checked and applied with `git apply`). Skipping auto-fix. |
| `msg_autofix_dirty_tree` | ⚠️ Working tree has uncommitted changes outside the scan scope. Recommend committing or backing up before running `--auto-fix`. |
| `msg_autofix_snapshot_scope` | ℹ️ Scope `commit id`/`pr id` scans a temporary snapshot, not your working tree — auto-fix only writes patch suggestions to `vbsec-reports/patches/` and does not modify any files. |
| `msg_autofix_staged_scope` | ℹ️ Scope `staged`: applying a patch only changes the working tree, not the index, so your next commit would still contain the issue. Auto-fix only writes patch suggestions to `vbsec-reports/patches/`; apply the ones you want with `git apply`, then `git add`. |
| `msg_autofix_needs_run_tests` | ℹ️ Dependency upgrades are only verified by running your project's tests, which can touch real systems (databases, services from `.env`). Add `--run-tests` to allow it; for now they are patch suggestions only. |
| `autofix_status_applied` | Fixed and build verification passed |
| `autofix_status_failed` | Could not fix after {n} attempts — needs manual fix |
| `autofix_status_suggested` | Patch suggestion only (no build command to verify) — see `vbsec-reports/patches/` |
| `autofix_status_skipped` | Skipped (MEDIUM/LOW severity, not auto-fixed) |
| `msg_autofix_summary` | Auto-fixed {applied}/{total} CRITICAL+HIGH issues. {failed} need manual fixes, {suggested} have patch suggestions only. |
## SCA / Live CVE (`--sca`, v0.7+)
| Key | Text |
|---|---|
| `header_sca_title` | Dependency scan (SCA — OSV.dev live) |
| `msg_sca_unavailable` | ⚠️ Could not reach OSV.dev — falling back to the offline static list (OUTDATED-DEPENDENCY rule). |
| `msg_sca_no_manifest` | No dependency manifest found for {lang} — skipping SCA for this language. |
| `col_cve` | CVE |
| `col_installed_fixed` | Installed version → Fixed version |
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| rea-analysis-optimizationmorluto/reaOptimize REA providers and shared analysis workflows for agent context, latency, CPU, and memory while preserving correctness and Evidence. Use for analysis bottlenecks and cache or query design; CI/test-suite optimization belongs to rea-ci-tests-audit. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| rea-ci-tests-auditmorluto/reaAudit REA CI and test suites for correctness, fixture drift, false passes, and measured performance; use for suite or workflow investigations and efficiency changes, rather than ordinary feature testing or a single test failure. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| rea-tool-designmorluto/reaAudit, investigate, design, or change REA investigation tools, CLI/MCP contracts, provider capabilities, and Evidence semantics. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-auditDietrichGebert/ponytailQuality audit of the whole repo: bugs, security, real load, missing tests, speed, and what to delete. Most important first. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |