anvil-engineering-workbench 1.0.6 → 1.0.8

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (38) hide show
  1. package/CHANGELOG.md +47 -2
  2. package/README.md +41 -35
  3. package/SECURITY.md +54 -10
  4. package/dist/ai/claude-code-provider.js +4 -3
  5. package/dist/ai/claude-code-provider.js.map +1 -1
  6. package/dist/ai/provider.js +7 -6
  7. package/dist/ai/provider.js.map +1 -1
  8. package/dist/cli.js +273 -59
  9. package/dist/cli.js.map +1 -1
  10. package/dist/core/adr-engine.d.ts +30 -0
  11. package/dist/core/adr-engine.js +163 -0
  12. package/dist/core/adr-engine.js.map +1 -0
  13. package/dist/core/code-review.js +6 -6
  14. package/dist/core/container-sandbox.js +1 -1
  15. package/dist/core/container-sandbox.js.map +1 -1
  16. package/dist/core/dashboard-server.js +222 -14
  17. package/dist/core/dashboard-server.js.map +1 -1
  18. package/dist/core/env-compat.d.ts +11 -0
  19. package/dist/core/env-compat.js +32 -0
  20. package/dist/core/env-compat.js.map +1 -0
  21. package/dist/core/hosted-agent-provider.js +11 -6
  22. package/dist/core/hosted-agent-provider.js.map +1 -1
  23. package/dist/core/project-cleaner.d.ts +7 -0
  24. package/dist/core/project-cleaner.js +258 -19
  25. package/dist/core/project-cleaner.js.map +1 -1
  26. package/dist/core/provider-diagnostics.js +3 -3
  27. package/dist/core/provider-diagnostics.js.map +1 -1
  28. package/dist/core/sarif-formatter.d.ts +12 -0
  29. package/dist/core/sarif-formatter.js +121 -0
  30. package/dist/core/sarif-formatter.js.map +1 -0
  31. package/dist/core/security-scanner.d.ts +37 -0
  32. package/dist/core/security-scanner.js +325 -0
  33. package/dist/core/security-scanner.js.map +1 -0
  34. package/dist/core/verification.js +2 -1
  35. package/dist/core/verification.js.map +1 -1
  36. package/dist/mcp/server.js +52 -32
  37. package/dist/mcp/server.js.map +1 -1
  38. package/package.json +9 -4
package/dist/cli.js CHANGED
@@ -2,7 +2,9 @@
2
2
  import path from "node:path";
3
3
  import { spawn } from "node:child_process";
4
4
  import os from "node:os";
5
- import { mkdtemp, rm } from "node:fs/promises";
5
+ import { mkdtemp, rm, mkdir, writeFile } from "node:fs/promises";
6
+ import { getAnvilEnv } from "./core/env-compat.js";
7
+ import { generateSarif } from "./core/sarif-formatter.js";
6
8
  import { initializeProject } from "./core/project-init.js";
7
9
  import { formatArchitectureMap, mapArchitecture } from "./core/architecture-map.js";
8
10
  import { formatProjectReport, inspectProject } from "./core/project-inspector.js";
@@ -46,6 +48,8 @@ import { analyzeDependencies, formatDependencyIntelligence } from "./core/depend
46
48
  import { generateRemediationPlan, formatRemediationPlan } from "./core/remediation-planner.js";
47
49
  import { addProjectMemory, buildProjectMemoryContext, listProjectMemory, removeProjectMemory } from "./core/project-memory.js";
48
50
  import { setupHarnessMcp, resolveMcpServerScript } from "./core/mcp-setup.js";
51
+ import { initAdrDirectory, listAdrs, createAdr, syncMemoryToAdrs } from "./core/adr-engine.js";
52
+ import { scanProjectSecurity, applySecurityFixes, formatSecurityReport } from "./core/security-scanner.js";
49
53
  import { formatSkillContext, loadSkillRegistry, selectSkills } from "./core/skill-registry.js";
50
54
  import { applyBeforeAgentHooks, loadHooks } from "./core/hook-engine.js";
51
55
  import { assertSupportedAgentPolicy, loadPermissionPolicy } from "./core/permission-policy.js";
@@ -97,7 +101,7 @@ function runSandboxProbe(args) {
97
101
  });
98
102
  }
99
103
  function readAgentTokenBudgets() {
100
- const raw = process.env.ANVIL_MAX_ESTIMATED_TOKENS_PER_AGENT ?? process.env.FORGE_MAX_ESTIMATED_TOKENS_PER_AGENT;
104
+ const raw = getAnvilEnv("ANVIL_MAX_ESTIMATED_TOKENS_PER_AGENT", "FORGE_MAX_ESTIMATED_TOKENS_PER_AGENT");
101
105
  if (!raw)
102
106
  return undefined;
103
107
  const parsed = JSON.parse(raw);
@@ -112,7 +116,7 @@ function readAgentTokenBudgets() {
112
116
  return budgets;
113
117
  }
114
118
  function readAgentDurationBudgets() {
115
- const raw = process.env.ANVIL_MAX_TASK_DURATION_MS_BY_AGENT ?? process.env.FORGE_MAX_TASK_DURATION_MS_BY_AGENT;
119
+ const raw = getAnvilEnv("ANVIL_MAX_TASK_DURATION_MS_BY_AGENT", "FORGE_MAX_TASK_DURATION_MS_BY_AGENT");
116
120
  if (!raw)
117
121
  return undefined;
118
122
  const parsed = JSON.parse(raw);
@@ -126,7 +130,7 @@ function readAgentDurationBudgets() {
126
130
  }
127
131
  return budgets;
128
132
  }
129
- const VERSION = "1.0.6";
133
+ const VERSION = "1.0.7";
130
134
  const HELP = `
131
135
  ANVIL — The AI Engineering Workbench
132
136
 
@@ -149,7 +153,7 @@ Commands:
149
153
  agent-release <prepare|list|show|approve|reject> [options] Gate agent configuration promotion
150
154
  experiment <capture|show|compare> [options] Capture and compare reproducible benchmark manifests
151
155
  quality [path] Generate an evidence-based static engineering quality report
152
- clean [path] Clean farzi/junk files and logs; use --apply to delete, --fix to repair code
156
+ clean [path] Clean workspace junk files and logs; use --apply to delete, --fix to repair code
153
157
  fix [path] Auto-fix code issues, unused imports, and formatting
154
158
  deps [path] Analyze npm dependencies, licenses, and imported security reports
155
159
  remediate [path] Advisory remediation plan for vulnerable and outdated dependencies
@@ -185,10 +189,13 @@ Commands:
185
189
  map [path] Map source modules and import relationships
186
190
  agents [path] List built-in and workspace agent roles
187
191
  memory <action> Manage local project memory (add, list, remove, context)
192
+ adr <action> Manage Architecture Decision Records (init, new, list, sync)
188
193
  mcp [setup] Model Context Protocol server and harness setup (cursor, claude, windsurf)
189
194
  skills [path] List built-in and workspace skills
190
195
  hooks [path] List configured declarative lifecycle hooks
191
196
  agent-security [path] Audit agent instructions, skills, hooks, and permissions
197
+ security [path] Audit repository for hardcoded secrets, keys, and script vulnerabilities
198
+ gate [path] Deterministic CI quality & security gate with SARIF and policy enforcement
192
199
  plan --goal TEXT Create and save a reviewable draft plan
193
200
  run --goal TEXT Run the plan using a configured AI provider
194
201
  runs [path] List locally saved run history (add --json for versioned JSON)
@@ -592,8 +599,8 @@ async function main(args) {
592
599
  const matrixFile = option("--matrix-file");
593
600
  const suiteId = option("--suite");
594
601
  const matrixId = option("--matrix-id");
595
- const provider = option("--provider") ?? process.env.FORGE_AI_PROVIDER ?? "openai";
596
- const model = option("--model") ?? process.env.FORGE_AI_MODEL ?? process.env.FORGE_MODEL ?? "gpt-4.1";
602
+ const provider = option("--provider") ?? getAnvilEnv("ANVIL_AI_PROVIDER", "FORGE_AI_PROVIDER", "openai");
603
+ const model = option("--model") ?? getAnvilEnv("ANVIL_AI_MODEL", "FORGE_AI_MODEL") ?? getAnvilEnv("ANVIL_MODEL", "FORGE_MODEL", "gpt-4.1");
597
604
  if (!matrixFile || !suiteId || !matrixId)
598
605
  throw new Error("Usage: anvil experiment capture --matrix-file CONFIG.json --matrix-id ID --suite SUITE_ID [--provider ID] [--model NAME] [--path PROJECT]");
599
606
  const matrix = await loadBenchmarkMatrix(path.resolve(matrixFile));
@@ -608,7 +615,8 @@ async function main(args) {
608
615
  const snapshot = await loadAgentConfigVersion(target, candidate.configVersionId);
609
616
  fingerprints.push({ candidateId: candidate.id, configVersionId: snapshot.id, fingerprint: fingerprintAgentConfig(snapshot.files) });
610
617
  }
611
- const manifest = await captureExperimentManifest({ projectPath: target, matrixFile: path.resolve(matrixFile), matrixId, suiteId, suite, provider, model, ...(process.env.FORGE_BASE_URL ? { baseUrl: process.env.FORGE_BASE_URL } : {}), agentConfigFingerprints: fingerprints });
618
+ const baseUrl = getAnvilEnv("ANVIL_BASE_URL", "FORGE_BASE_URL");
619
+ const manifest = await captureExperimentManifest({ projectPath: target, matrixFile: path.resolve(matrixFile), matrixId, suiteId, suite, provider, model, ...(baseUrl ? { baseUrl } : {}), agentConfigFingerprints: fingerprints });
612
620
  const saved = await saveExperimentManifest(target, manifest);
613
621
  process.stdout.write(formatExperimentManifest(manifest) + "\nSaved: " + path.relative(target, saved) + "\n");
614
622
  }
@@ -662,22 +670,22 @@ async function main(args) {
662
670
  const suite = suites.find(item => item.id === suiteId);
663
671
  if (!suite)
664
672
  throw new Error("Evaluation suite not found: " + suiteId);
665
- const apiKey = process.env.FORGE_API_KEY ?? process.env.OPENAI_API_KEY ?? "";
666
- const baseUrl = process.env.FORGE_BASE_URL ?? "https://api.openai.com/v1";
673
+ const apiKey = getAnvilEnv("ANVIL_API_KEY", "FORGE_API_KEY") ?? process.env.OPENAI_API_KEY ?? "";
674
+ const baseUrl = getAnvilEnv("ANVIL_BASE_URL", "FORGE_BASE_URL", "https://api.openai.com/v1");
667
675
  const report = await executeBenchmarkMatrix(config, async (candidate, _repeat) => {
668
676
  const createProvider = (selectedModel) => candidate.provider === "claude-code"
669
677
  ? new ClaudeCodeProvider({ cwd: target, model: selectedModel })
670
- : createHostedAgentProvider({ ...process.env, FORGE_AI_PROVIDER: candidate.provider, ...(apiKey ? { FORGE_API_KEY: apiKey } : {}), ...(process.env.FORGE_BASE_URL ? { FORGE_BASE_URL: process.env.FORGE_BASE_URL } : {}) }, selectedModel);
678
+ : createHostedAgentProvider({ ...process.env, ANVIL_AI_PROVIDER: candidate.provider, FORGE_AI_PROVIDER: candidate.provider, ...(apiKey ? { ANVIL_API_KEY: apiKey, FORGE_API_KEY: apiKey } : {}), ...(baseUrl ? { ANVIL_BASE_URL: baseUrl, FORGE_BASE_URL: baseUrl } : {}) }, selectedModel);
671
679
  const router = new ModelRouter({
672
680
  fastModel: candidate.model, balancedModel: candidate.model, reasoningModel: candidate.model,
673
- retryAttempts: Number(process.env.FORGE_MODEL_RETRY_ATTEMPTS ?? 2),
674
- retryDelayMs: Number(process.env.FORGE_MODEL_RETRY_DELAY_MS ?? 500),
675
- ...(process.env.FORGE_MAX_INPUT_TOKENS_PER_TASK ? { maxInputTokensPerTask: Number(process.env.FORGE_MAX_INPUT_TOKENS_PER_TASK) } : {}),
676
- ...(process.env.FORGE_MAX_OUTPUT_TOKENS_PER_TASK ? { maxOutputTokensPerTask: Number(process.env.FORGE_MAX_OUTPUT_TOKENS_PER_TASK) } : {}),
677
- ...(process.env.FORGE_MAX_ESTIMATED_TOKENS_PER_RUN ? { maxEstimatedTokensPerRun: Number(process.env.FORGE_MAX_ESTIMATED_TOKENS_PER_RUN) } : {}),
678
- ...(process.env.FORGE_MAX_ESTIMATED_TOKENS_PER_AGENT ? { maxEstimatedTokensPerAgent: readAgentTokenBudgets() } : {}),
679
- ...(process.env.FORGE_MAX_TASK_DURATION_MS ? { maxTaskDurationMs: Number(process.env.FORGE_MAX_TASK_DURATION_MS) } : {}),
680
- ...(process.env.FORGE_MAX_TASK_DURATION_MS_BY_AGENT ? { maxTaskDurationMsByAgent: readAgentDurationBudgets() } : {}),
681
+ retryAttempts: Number(getAnvilEnv("ANVIL_MODEL_RETRY_ATTEMPTS", "FORGE_MODEL_RETRY_ATTEMPTS", "2")),
682
+ retryDelayMs: Number(getAnvilEnv("ANVIL_MODEL_RETRY_DELAY_MS", "FORGE_MODEL_RETRY_DELAY_MS", "500")),
683
+ ...(getAnvilEnv("ANVIL_MAX_INPUT_TOKENS_PER_TASK", "FORGE_MAX_INPUT_TOKENS_PER_TASK") ? { maxInputTokensPerTask: Number(getAnvilEnv("ANVIL_MAX_INPUT_TOKENS_PER_TASK", "FORGE_MAX_INPUT_TOKENS_PER_TASK")) } : {}),
684
+ ...(getAnvilEnv("ANVIL_MAX_OUTPUT_TOKENS_PER_TASK", "FORGE_MAX_OUTPUT_TOKENS_PER_TASK") ? { maxOutputTokensPerTask: Number(getAnvilEnv("ANVIL_MAX_OUTPUT_TOKENS_PER_TASK", "FORGE_MAX_OUTPUT_TOKENS_PER_TASK")) } : {}),
685
+ ...(getAnvilEnv("ANVIL_MAX_ESTIMATED_TOKENS_PER_RUN", "FORGE_MAX_ESTIMATED_TOKENS_PER_RUN") ? { maxEstimatedTokensPerRun: Number(getAnvilEnv("ANVIL_MAX_ESTIMATED_TOKENS_PER_RUN", "FORGE_MAX_ESTIMATED_TOKENS_PER_RUN")) } : {}),
686
+ ...(getAnvilEnv("ANVIL_MAX_ESTIMATED_TOKENS_PER_AGENT", "FORGE_MAX_ESTIMATED_TOKENS_PER_AGENT") ? { maxEstimatedTokensPerAgent: readAgentTokenBudgets() } : {}),
687
+ ...(getAnvilEnv("ANVIL_MAX_TASK_DURATION_MS", "FORGE_MAX_TASK_DURATION_MS") ? { maxTaskDurationMs: Number(getAnvilEnv("ANVIL_MAX_TASK_DURATION_MS", "FORGE_MAX_TASK_DURATION_MS")) } : {}),
688
+ ...(getAnvilEnv("ANVIL_MAX_TASK_DURATION_MS_BY_AGENT", "FORGE_MAX_TASK_DURATION_MS_BY_AGENT") ? { maxTaskDurationMsByAgent: readAgentDurationBudgets() } : {}),
681
689
  createProvider,
682
690
  });
683
691
  const configWorkspace = await mkdtemp(path.join(os.tmpdir(), "forge-candidate-config-"));
@@ -794,12 +802,12 @@ async function main(args) {
794
802
  const judgeEnabled = rest.includes("--judge");
795
803
  let report;
796
804
  if (judgeEnabled) {
797
- const providerChoice = process.env.FORGE_AI_PROVIDER ?? "openai";
798
- const model = process.env.FORGE_EVAL_JUDGE_MODEL ?? process.env.FORGE_AI_MODEL ?? process.env.FORGE_MODEL ?? (providerChoice === "claude-code" ? "sonnet" : "gpt-4.1");
799
- const apiKey = process.env.FORGE_API_KEY ?? process.env.OPENAI_API_KEY ?? "";
800
- const baseUrl = process.env.FORGE_BASE_URL ?? "https://api.openai.com/v1";
805
+ const providerChoice = getAnvilEnv("ANVIL_AI_PROVIDER", "FORGE_AI_PROVIDER", "openai");
806
+ const model = getAnvilEnv("ANVIL_EVAL_JUDGE_MODEL", "FORGE_EVAL_JUDGE_MODEL") ?? getAnvilEnv("ANVIL_AI_MODEL", "FORGE_AI_MODEL") ?? getAnvilEnv("ANVIL_MODEL", "FORGE_MODEL") ?? (providerChoice === "claude-code" ? "sonnet" : "gpt-4.1");
807
+ const apiKey = getAnvilEnv("ANVIL_API_KEY", "FORGE_API_KEY") ?? process.env.OPENAI_API_KEY ?? "";
808
+ const baseUrl = getAnvilEnv("ANVIL_BASE_URL", "FORGE_BASE_URL", "https://api.openai.com/v1");
801
809
  if (providerChoice !== "claude-code" && !apiKey)
802
- throw new Error("Set FORGE_API_KEY or OPENAI_API_KEY to use model-judged evaluation.");
810
+ throw new Error("Set ANVIL_API_KEY, FORGE_API_KEY, or OPENAI_API_KEY to use model-judged evaluation.");
803
811
  const judgeProvider = providerChoice === "claude-code" ? new ClaudeCodeProvider({ cwd: target, model }) : new OpenAICompatibleProvider({ apiKey, model, baseUrl });
804
812
  report = await evaluateRunWithJudge(run, suite, judgeProvider);
805
813
  }
@@ -925,7 +933,8 @@ async function main(args) {
925
933
  skipValues.add(i + 1);
926
934
  }
927
935
  const positional = memoryArgs.filter((arg, i) => !arg.startsWith("--") && !skipValues.has(i));
928
- const target = path.resolve(positional[0] ?? ".");
936
+ const targetPath = positional.find(arg => !arg.startsWith("mem-")) ?? ".";
937
+ const target = path.resolve(targetPath);
929
938
  try {
930
939
  if (action === "list") {
931
940
  const entries = await listProjectMemory(target);
@@ -1004,6 +1013,207 @@ async function main(args) {
1004
1013
  }
1005
1014
  return;
1006
1015
  }
1016
+ case "adr": {
1017
+ if (rest.includes("--help") || rest.includes("-h") || rest[0] === "help") {
1018
+ process.stdout.write(`Usage: anvil adr <subcommand> [options]
1019
+
1020
+ Manage Architecture Decision Records (ADR) using the MADR 3.0 standard.
1021
+
1022
+ Subcommands:
1023
+ init [path] Initialize docs/adr/ directory with template
1024
+ new "Title" [path] [--status STATUS] Create a new ADR record
1025
+ list [path] List all recorded ADRs
1026
+ sync [path] Sync decisions from project memory to docs/adr/
1027
+ `);
1028
+ return;
1029
+ }
1030
+ const [action = "list", ...adrArgs] = rest;
1031
+ const option = (name) => { const index = adrArgs.indexOf(name); return index >= 0 ? adrArgs[index + 1] : undefined; };
1032
+ const skipValues = new Set();
1033
+ for (let i = 0; i < adrArgs.length; i++) {
1034
+ if (adrArgs[i]?.startsWith("--"))
1035
+ skipValues.add(i + 1);
1036
+ }
1037
+ const positional = adrArgs.filter((arg, i) => !arg.startsWith("--") && !skipValues.has(i));
1038
+ try {
1039
+ if (action === "init") {
1040
+ const target = path.resolve(positional[0] ?? ".");
1041
+ const res = await initAdrDirectory(target);
1042
+ process.stdout.write(res.created ? `Initialized Architecture Decision Records in ${res.path}\n` : `ADR directory already initialized in ${res.path}\n`);
1043
+ }
1044
+ else if (action === "new") {
1045
+ const title = positional[0];
1046
+ if (!title)
1047
+ throw new Error("Usage: anvil adr new \"Title of decision\" [path] [--status Proposed|Accepted]");
1048
+ const target = path.resolve(positional[1] ?? ".");
1049
+ const status = (option("--status") ?? "Accepted");
1050
+ const context = option("--context");
1051
+ const decision = option("--decision");
1052
+ const record = await createAdr(target, title, { status, context, decision });
1053
+ process.stdout.write(`Created ADR ${record.id}: ${record.filename}\n`);
1054
+ }
1055
+ else if (action === "sync") {
1056
+ const target = path.resolve(positional[0] ?? ".");
1057
+ const res = await syncMemoryToAdrs(target);
1058
+ process.stdout.write(res.syncedCount > 0
1059
+ ? `Synced ${res.syncedCount} decisions from project memory to docs/adr/:\n` + res.createdAdrs.map(a => ` - [${a.id}] ${a.filename}`).join("\n") + "\n"
1060
+ : `All project memory decisions are already synced to docs/adr/.\n`);
1061
+ }
1062
+ else if (action === "list") {
1063
+ const target = path.resolve(positional[0] ?? ".");
1064
+ const records = await listAdrs(target);
1065
+ if (!records.length) {
1066
+ process.stdout.write("No Architecture Decision Records found. Run 'anvil adr init' to get started.\n");
1067
+ }
1068
+ else {
1069
+ process.stdout.write(`Architecture Decision Records (${records.length}):\n\n`);
1070
+ for (const r of records) {
1071
+ process.stdout.write(` [${r.id}] ${r.title} (${r.status}, ${r.date})\n`);
1072
+ }
1073
+ }
1074
+ }
1075
+ else {
1076
+ throw new Error("ADR actions: init, new, list, sync.");
1077
+ }
1078
+ }
1079
+ catch (error) {
1080
+ process.stderr.write("ANVIL ADR command failed: " + (error instanceof Error ? error.message : String(error)) + "\n");
1081
+ process.exitCode = 1;
1082
+ }
1083
+ return;
1084
+ }
1085
+ case "security": {
1086
+ if (rest.includes("--help") || rest.includes("-h") || rest[0] === "help") {
1087
+ process.stdout.write(`Usage: anvil security [path] [options]
1088
+
1089
+ Deep scan repository for hardcoded secrets, exposed tokens, sensitive files, and dangerous script patterns.
1090
+
1091
+ Options:
1092
+ --fix Auto-remediate exposed credentials by updating .gitignore
1093
+ --json Output report in machine-readable JSON format
1094
+ --help Show this help message
1095
+ `);
1096
+ return;
1097
+ }
1098
+ const isJson = rest.includes("--json");
1099
+ const shouldFix = rest.includes("--fix");
1100
+ const positional = rest.filter(arg => !arg.startsWith("--"));
1101
+ const target = path.resolve(positional[0] ?? ".");
1102
+ try {
1103
+ const report = await scanProjectSecurity(target);
1104
+ if (shouldFix && report.findings.length > 0) {
1105
+ const fixResult = await applySecurityFixes(target, report.findings);
1106
+ if (fixResult.fixedCount > 0) {
1107
+ process.stdout.write(`Applied ${fixResult.fixedCount} security fix(es):\n` + fixResult.details.map(d => ` ✓ ${d}`).join("\n") + "\n\n");
1108
+ }
1109
+ }
1110
+ if (isJson) {
1111
+ process.stdout.write(JSON.stringify(report, null, 2) + "\n");
1112
+ }
1113
+ else {
1114
+ process.stdout.write(formatSecurityReport(report));
1115
+ }
1116
+ if (!report.passed) {
1117
+ process.exitCode = 1;
1118
+ }
1119
+ }
1120
+ catch (error) {
1121
+ process.stderr.write("ANVIL security audit failed: " + (error instanceof Error ? error.message : String(error)) + "\n");
1122
+ process.exitCode = 1;
1123
+ }
1124
+ return;
1125
+ }
1126
+ case "gate": {
1127
+ if (rest.includes("--help") || rest.includes("-h") || rest[0] === "help") {
1128
+ process.stdout.write(`Usage: anvil gate [path] [options]
1129
+
1130
+ Deterministic CI quality & security gate. Evaluates repository health, security secrets,
1131
+ and code quality rules against policies. Emits human-readable reports, machine-readable JSON,
1132
+ or OASIS SARIF v2.1.0 for PR comments and GitHub Security tab integration.
1133
+
1134
+ Options:
1135
+ --sarif PATH Export findings to an OASIS SARIF v2.1.0 report file
1136
+ --policy PATH Custom JSON policy file (defaults to .anvil/policy.json if present)
1137
+ --fail-on-warnings Fail gate (exit 1) on medium-severity warnings as well as critical/high errors
1138
+ --json Output full gate evaluation in machine-readable JSON format
1139
+ --help Show this help message
1140
+
1141
+ Exit codes:
1142
+ 0 Gate passed: all policy thresholds and security checks satisfied.
1143
+ 1 Gate failed: critical/high vulnerabilities or policy violations detected.
1144
+ `);
1145
+ return;
1146
+ }
1147
+ const option = (name) => {
1148
+ const index = rest.indexOf(name);
1149
+ if (index < 0)
1150
+ return undefined;
1151
+ const next = rest[index + 1];
1152
+ return next && !next.startsWith("--") ? next : undefined;
1153
+ };
1154
+ const isJson = rest.includes("--json");
1155
+ const failOnWarnings = rest.includes("--fail-on-warnings");
1156
+ const sarifPath = option("--sarif");
1157
+ const policyPath = option("--policy");
1158
+ const positional = rest.filter(arg => !arg.startsWith("--") && arg !== sarifPath && arg !== policyPath);
1159
+ const target = path.resolve(positional[0] ?? ".");
1160
+ try {
1161
+ const securityReport = await scanProjectSecurity(target);
1162
+ const reviewReport = await reviewProject(target, { all: true });
1163
+ if (sarifPath) {
1164
+ const sarifOutput = generateSarif({ securityReport, reviewReport, targetRoot: target });
1165
+ const outResolved = path.resolve(sarifPath);
1166
+ await mkdir(path.dirname(outResolved), { recursive: true });
1167
+ await writeFile(outResolved, sarifOutput, "utf8");
1168
+ if (!isJson) {
1169
+ process.stdout.write(`Exported SARIF report to ${path.relative(process.cwd(), outResolved)}\n`);
1170
+ }
1171
+ }
1172
+ const criticalCount = securityReport.summary.critical;
1173
+ const highCount = securityReport.summary.high + reviewReport.summary.high;
1174
+ const mediumCount = securityReport.summary.medium + reviewReport.summary.medium;
1175
+ const passed = criticalCount === 0 && highCount === 0 && (!failOnWarnings || mediumCount === 0);
1176
+ if (isJson) {
1177
+ process.stdout.write(JSON.stringify({
1178
+ gate: passed ? "PASSED" : "FAILED",
1179
+ target,
1180
+ summary: {
1181
+ critical: criticalCount,
1182
+ high: highCount,
1183
+ medium: mediumCount,
1184
+ low: securityReport.summary.low + reviewReport.summary.low
1185
+ },
1186
+ security: securityReport,
1187
+ review: reviewReport
1188
+ }, null, 2) + "\n");
1189
+ }
1190
+ else {
1191
+ process.stdout.write([
1192
+ `# ANVIL CI Quality & Security Gate`,
1193
+ `Target: ${target}`,
1194
+ `Status: ${passed ? "PASSED" : "FAILED"}`,
1195
+ ``,
1196
+ `Findings Summary:`,
1197
+ ` Critical: ${criticalCount}`,
1198
+ ` High: ${highCount}`,
1199
+ ` Medium: ${mediumCount}`,
1200
+ ` Low: ${securityReport.summary.low + reviewReport.summary.low}`,
1201
+ ``,
1202
+ passed
1203
+ ? `✓ Gate passed: 0 critical/high findings detected.`
1204
+ : `✗ Gate failed: ${criticalCount + highCount} critical or high findings require remediation before merging.`
1205
+ ].join("\n") + "\n");
1206
+ }
1207
+ if (!passed) {
1208
+ process.exitCode = 1;
1209
+ }
1210
+ }
1211
+ catch (error) {
1212
+ process.stderr.write("ANVIL gate check failed: " + (error instanceof Error ? error.message : String(error)) + "\n");
1213
+ process.exitCode = 1;
1214
+ }
1215
+ return;
1216
+ }
1007
1217
  case "plan": {
1008
1218
  const goalIndex = rest.indexOf("--goal");
1009
1219
  const pathIndex = rest.indexOf("--path");
@@ -1071,21 +1281,21 @@ async function main(args) {
1071
1281
  if (!runId)
1072
1282
  throw new Error("Usage: anvil resume RUN_ID [--path PROJECT]");
1073
1283
  const permissions = await loadPermissionPolicy(target);
1074
- const providerChoice = (process.env.FORGE_AI_PROVIDER ?? "openai").toLowerCase();
1075
- assertSupportedAgentPolicy(permissions, providerChoice, "FORGE resume");
1284
+ const providerChoice = (getAnvilEnv("ANVIL_AI_PROVIDER", "FORGE_AI_PROVIDER", "openai")).toLowerCase();
1285
+ assertSupportedAgentPolicy(permissions, providerChoice, "ANVIL resume");
1076
1286
  await withRunExecutionLock(target, runId, async () => {
1077
1287
  const savedRun = await loadRun(target, runId);
1078
1288
  if (savedRun.status === "running") {
1079
1289
  const recovered = await recoverInterruptedRun(target, runId);
1080
1290
  process.stdout.write("Recovered interrupted run " + runId + ": " + recovered.status + "\n");
1081
- process.stdout.write("Any task that was running has an unknown provider outcome and was marked failed. FORGE did not replay provider requests. Review the run before explicitly retrying failed work.\n");
1291
+ process.stdout.write("Any task that was running has an unknown provider outcome and was marked failed. ANVIL did not replay provider requests. Review the run before explicitly retrying failed work.\n");
1082
1292
  return;
1083
1293
  }
1084
1294
  if (savedRun.status !== "pending" && savedRun.status !== "awaiting_approval")
1085
- throw new Error("Only pending or approval-cleared runs can be resumed. Use `forge replay` only after reviewing failed tasks.");
1086
- const model = process.env.FORGE_AI_MODEL ?? process.env.FORGE_MODEL;
1295
+ throw new Error("Only pending or approval-cleared runs can be resumed. Use `anvil replay` only after reviewing failed tasks.");
1296
+ const model = getAnvilEnv("ANVIL_AI_MODEL", "FORGE_AI_MODEL") ?? getAnvilEnv("ANVIL_MODEL", "FORGE_MODEL");
1087
1297
  const baseModel = model ?? (providerChoice === "claude-code" ? "sonnet" : "gpt-4.1-mini");
1088
- const createProvider = (selectedModel) => providerChoice === "claude-code" ? new ClaudeCodeProvider({ cwd: target, model: selectedModel }) : createHostedAgentProvider({ ...process.env, FORGE_AI_PROVIDER: providerChoice }, selectedModel, { workspacePath: target, permissionPolicy: permissions });
1298
+ const createProvider = (selectedModel) => providerChoice === "claude-code" ? new ClaudeCodeProvider({ cwd: target, model: selectedModel }) : createHostedAgentProvider({ ...process.env, ANVIL_AI_PROVIDER: providerChoice, FORGE_AI_PROVIDER: providerChoice }, selectedModel, { workspacePath: target, permissionPolicy: permissions });
1089
1299
  let priorTraceEntries = [];
1090
1300
  try {
1091
1301
  priorTraceEntries = (await loadExecutionTrace(target, runId)).entries;
@@ -1098,11 +1308,14 @@ async function main(args) {
1098
1308
  const traceRunUsage = priorTraceEntries.reduce((total, entry) => total + entry.inputTokensEstimate + entry.outputTokensEstimate, 0);
1099
1309
  const priorAgentUsage = savedRun.tokenUsage?.estimatedTokensByAgent ?? traceAgentUsage;
1100
1310
  const priorRunUsage = savedRun.tokenUsage?.estimatedTokensUsed ?? traceRunUsage;
1101
- const router = new ModelRouter({ fastModel: process.env.FORGE_MODEL_FAST ?? baseModel, balancedModel: process.env.FORGE_MODEL_BALANCED ?? baseModel, reasoningModel: process.env.FORGE_MODEL_REASONING ?? baseModel, ...(process.env.FORGE_FALLBACK_MODEL ? { fallbackModel: process.env.FORGE_FALLBACK_MODEL } : {}), retryAttempts: Number(process.env.FORGE_MODEL_RETRY_ATTEMPTS ?? 2), retryDelayMs: Number(process.env.FORGE_MODEL_RETRY_DELAY_MS ?? 500), ...(process.env.FORGE_MAX_INPUT_TOKENS_PER_TASK ? { maxInputTokensPerTask: Number(process.env.FORGE_MAX_INPUT_TOKENS_PER_TASK) } : {}), ...(process.env.FORGE_MAX_OUTPUT_TOKENS_PER_TASK ? { maxOutputTokensPerTask: Number(process.env.FORGE_MAX_OUTPUT_TOKENS_PER_TASK) } : {}), ...(process.env.FORGE_MAX_ESTIMATED_TOKENS_PER_RUN ? { maxEstimatedTokensPerRun: Number(process.env.FORGE_MAX_ESTIMATED_TOKENS_PER_RUN) } : {}),
1102
- ...(process.env.FORGE_MAX_ESTIMATED_TOKENS_PER_AGENT ? { maxEstimatedTokensPerAgent: readAgentTokenBudgets() } : {}),
1103
- ...(process.env.FORGE_MAX_TASK_DURATION_MS ? { maxTaskDurationMs: Number(process.env.FORGE_MAX_TASK_DURATION_MS) } : {}),
1104
- ...(process.env.FORGE_MAX_TASK_DURATION_MS_BY_AGENT ? { maxTaskDurationMsByAgent: readAgentDurationBudgets() } : {}),
1105
- initialEstimatedTokensUsed: priorRunUsage, initialEstimatedTokensByAgent: priorAgentUsage, createProvider });
1311
+ const resumeFallback = getAnvilEnv("ANVIL_FALLBACK_MODEL", "FORGE_FALLBACK_MODEL");
1312
+ const router = new ModelRouter({
1313
+ fastModel: getAnvilEnv("ANVIL_MODEL_FAST", "FORGE_MODEL_FAST") ?? baseModel, balancedModel: getAnvilEnv("ANVIL_MODEL_BALANCED", "FORGE_MODEL_BALANCED") ?? baseModel, reasoningModel: getAnvilEnv("ANVIL_MODEL_REASONING", "FORGE_MODEL_REASONING") ?? baseModel, ...(resumeFallback ? { fallbackModel: resumeFallback } : {}), retryAttempts: Number(getAnvilEnv("ANVIL_MODEL_RETRY_ATTEMPTS", "FORGE_MODEL_RETRY_ATTEMPTS", "2")), retryDelayMs: Number(getAnvilEnv("ANVIL_MODEL_RETRY_DELAY_MS", "FORGE_MODEL_RETRY_DELAY_MS", "500")), ...(getAnvilEnv("ANVIL_MAX_INPUT_TOKENS_PER_TASK", "FORGE_MAX_INPUT_TOKENS_PER_TASK") ? { maxInputTokensPerTask: Number(getAnvilEnv("ANVIL_MAX_INPUT_TOKENS_PER_TASK", "FORGE_MAX_INPUT_TOKENS_PER_TASK")) } : {}), ...(getAnvilEnv("ANVIL_MAX_OUTPUT_TOKENS_PER_TASK", "FORGE_MAX_OUTPUT_TOKENS_PER_TASK") ? { maxOutputTokensPerTask: Number(getAnvilEnv("ANVIL_MAX_OUTPUT_TOKENS_PER_TASK", "FORGE_MAX_OUTPUT_TOKENS_PER_TASK")) } : {}), ...(getAnvilEnv("ANVIL_MAX_ESTIMATED_TOKENS_PER_RUN", "FORGE_MAX_ESTIMATED_TOKENS_PER_RUN") ? { maxEstimatedTokensPerRun: Number(getAnvilEnv("ANVIL_MAX_ESTIMATED_TOKENS_PER_RUN", "FORGE_MAX_ESTIMATED_TOKENS_PER_RUN")) } : {}),
1314
+ ...(getAnvilEnv("ANVIL_MAX_ESTIMATED_TOKENS_PER_AGENT", "FORGE_MAX_ESTIMATED_TOKENS_PER_AGENT") ? { maxEstimatedTokensPerAgent: readAgentTokenBudgets() } : {}),
1315
+ ...(getAnvilEnv("ANVIL_MAX_TASK_DURATION_MS", "FORGE_MAX_TASK_DURATION_MS") ? { maxTaskDurationMs: Number(getAnvilEnv("ANVIL_MAX_TASK_DURATION_MS", "FORGE_MAX_TASK_DURATION_MS")) } : {}),
1316
+ ...(getAnvilEnv("ANVIL_MAX_TASK_DURATION_MS_BY_AGENT", "FORGE_MAX_TASK_DURATION_MS_BY_AGENT") ? { maxTaskDurationMsByAgent: readAgentDurationBudgets() } : {}),
1317
+ initialEstimatedTokensUsed: priorRunUsage, initialEstimatedTokensByAgent: priorAgentUsage, createProvider
1318
+ });
1106
1319
  const memory = await buildProjectMemoryContext(target, savedRun.goal);
1107
1320
  // Re-check current workspace policy before resuming any task that has not completed.
1108
1321
  // Previously applied hook guidance is kept in the saved prompt and is not appended twice.
@@ -1263,34 +1476,35 @@ async function main(args) {
1263
1476
  const target = path.resolve(pathIndex >= 0 ? rest[pathIndex + 1] ?? "." : ".");
1264
1477
  try {
1265
1478
  const permissions = await loadPermissionPolicy(target);
1266
- const model = process.env.FORGE_AI_MODEL ?? process.env.FORGE_MODEL;
1267
- const providerChoice = (process.env.FORGE_AI_PROVIDER ?? "openai").toLowerCase();
1268
- assertSupportedAgentPolicy(permissions, providerChoice, "FORGE run");
1479
+ const model = getAnvilEnv("ANVIL_AI_MODEL", "FORGE_AI_MODEL") ?? getAnvilEnv("ANVIL_MODEL", "FORGE_MODEL");
1480
+ const providerChoice = (getAnvilEnv("ANVIL_AI_PROVIDER", "FORGE_AI_PROVIDER", "openai")).toLowerCase();
1481
+ assertSupportedAgentPolicy(permissions, providerChoice, "ANVIL run");
1269
1482
  const baseModel = model ?? (providerChoice === "claude-code" ? "sonnet" : "gpt-4.1-mini");
1270
- const apiKey = process.env.FORGE_API_KEY ?? process.env.OPENAI_API_KEY ?? "";
1271
- const baseUrl = process.env.FORGE_BASE_URL ?? "https://api.openai.com/v1";
1483
+ const apiKey = getAnvilEnv("ANVIL_API_KEY", "FORGE_API_KEY") ?? process.env.OPENAI_API_KEY ?? "";
1484
+ const baseUrl = getAnvilEnv("ANVIL_BASE_URL", "FORGE_BASE_URL", "https://api.openai.com/v1");
1272
1485
  const createProvider = (selectedModel) => {
1273
1486
  if (providerChoice === "claude-code")
1274
1487
  return new ClaudeCodeProvider({ cwd: target, model: selectedModel });
1275
1488
  if (["openai", "anthropic", "google", "ollama"].includes(providerChoice)) {
1276
- return createHostedAgentProvider({ ...process.env, FORGE_AI_PROVIDER: providerChoice }, selectedModel, { workspacePath: target, permissionPolicy: permissions });
1489
+ return createHostedAgentProvider({ ...process.env, ANVIL_AI_PROVIDER: providerChoice, FORGE_AI_PROVIDER: providerChoice }, selectedModel, { workspacePath: target, permissionPolicy: permissions });
1277
1490
  }
1278
- throw new Error("Unsupported FORGE_AI_PROVIDER. Choose claude-code, openai, anthropic, google or ollama.");
1491
+ throw new Error("Unsupported ANVIL_AI_PROVIDER. Choose claude-code, openai, anthropic, google or ollama.");
1279
1492
  };
1493
+ const runFallback = getAnvilEnv("ANVIL_FALLBACK_MODEL", "FORGE_FALLBACK_MODEL");
1280
1494
  const router = new ModelRouter({
1281
- fastModel: process.env.FORGE_MODEL_FAST ?? baseModel,
1282
- balancedModel: process.env.FORGE_MODEL_BALANCED ?? baseModel,
1283
- reasoningModel: process.env.FORGE_MODEL_REASONING ?? baseModel,
1284
- ...(process.env.FORGE_FALLBACK_MODEL ? { fallbackModel: process.env.FORGE_FALLBACK_MODEL } : {}),
1285
- retryAttempts: Number(process.env.FORGE_MODEL_RETRY_ATTEMPTS ?? 2),
1286
- retryDelayMs: Number(process.env.FORGE_MODEL_RETRY_DELAY_MS ?? 500),
1495
+ fastModel: getAnvilEnv("ANVIL_MODEL_FAST", "FORGE_MODEL_FAST") ?? baseModel,
1496
+ balancedModel: getAnvilEnv("ANVIL_MODEL_BALANCED", "FORGE_MODEL_BALANCED") ?? baseModel,
1497
+ reasoningModel: getAnvilEnv("ANVIL_MODEL_REASONING", "FORGE_MODEL_REASONING") ?? baseModel,
1498
+ ...(runFallback ? { fallbackModel: runFallback } : {}),
1499
+ retryAttempts: Number(getAnvilEnv("ANVIL_MODEL_RETRY_ATTEMPTS", "FORGE_MODEL_RETRY_ATTEMPTS", "2")),
1500
+ retryDelayMs: Number(getAnvilEnv("ANVIL_MODEL_RETRY_DELAY_MS", "FORGE_MODEL_RETRY_DELAY_MS", "500")),
1287
1501
  createProvider,
1288
- ...(process.env.FORGE_MAX_INPUT_TOKENS_PER_TASK ? { maxInputTokensPerTask: Number(process.env.FORGE_MAX_INPUT_TOKENS_PER_TASK) } : {}),
1289
- ...(process.env.FORGE_MAX_OUTPUT_TOKENS_PER_TASK ? { maxOutputTokensPerTask: Number(process.env.FORGE_MAX_OUTPUT_TOKENS_PER_TASK) } : {}),
1290
- ...(process.env.FORGE_MAX_ESTIMATED_TOKENS_PER_RUN ? { maxEstimatedTokensPerRun: Number(process.env.FORGE_MAX_ESTIMATED_TOKENS_PER_RUN) } : {}),
1291
- ...(process.env.FORGE_MAX_ESTIMATED_TOKENS_PER_AGENT ? { maxEstimatedTokensPerAgent: readAgentTokenBudgets() } : {}),
1292
- ...(process.env.FORGE_MAX_TASK_DURATION_MS ? { maxTaskDurationMs: Number(process.env.FORGE_MAX_TASK_DURATION_MS) } : {}),
1293
- ...(process.env.FORGE_MAX_TASK_DURATION_MS_BY_AGENT ? { maxTaskDurationMsByAgent: readAgentDurationBudgets() } : {}),
1502
+ ...(getAnvilEnv("ANVIL_MAX_INPUT_TOKENS_PER_TASK", "FORGE_MAX_INPUT_TOKENS_PER_TASK") ? { maxInputTokensPerTask: Number(getAnvilEnv("ANVIL_MAX_INPUT_TOKENS_PER_TASK", "FORGE_MAX_INPUT_TOKENS_PER_TASK")) } : {}),
1503
+ ...(getAnvilEnv("ANVIL_MAX_OUTPUT_TOKENS_PER_TASK", "FORGE_MAX_OUTPUT_TOKENS_PER_TASK") ? { maxOutputTokensPerTask: Number(getAnvilEnv("ANVIL_MAX_OUTPUT_TOKENS_PER_TASK", "FORGE_MAX_OUTPUT_TOKENS_PER_TASK")) } : {}),
1504
+ ...(getAnvilEnv("ANVIL_MAX_ESTIMATED_TOKENS_PER_RUN", "FORGE_MAX_ESTIMATED_TOKENS_PER_RUN") ? { maxEstimatedTokensPerRun: Number(getAnvilEnv("ANVIL_MAX_ESTIMATED_TOKENS_PER_RUN", "FORGE_MAX_ESTIMATED_TOKENS_PER_RUN")) } : {}),
1505
+ ...(getAnvilEnv("ANVIL_MAX_ESTIMATED_TOKENS_PER_AGENT", "FORGE_MAX_ESTIMATED_TOKENS_PER_AGENT") ? { maxEstimatedTokensPerAgent: readAgentTokenBudgets() } : {}),
1506
+ ...(getAnvilEnv("ANVIL_MAX_TASK_DURATION_MS", "FORGE_MAX_TASK_DURATION_MS") ? { maxTaskDurationMs: Number(getAnvilEnv("ANVIL_MAX_TASK_DURATION_MS", "FORGE_MAX_TASK_DURATION_MS")) } : {}),
1507
+ ...(getAnvilEnv("ANVIL_MAX_TASK_DURATION_MS_BY_AGENT", "FORGE_MAX_TASK_DURATION_MS_BY_AGENT") ? { maxTaskDurationMsByAgent: readAgentDurationBudgets() } : {}),
1294
1508
  });
1295
1509
  const provider = router;
1296
1510
  const plan = createReviewPlan(goal);
@@ -1731,10 +1945,10 @@ async function main(args) {
1731
1945
  const report = await reviewProject(target, { base, all });
1732
1946
  let ai;
1733
1947
  if (rest.includes("--ai")) {
1734
- const apiKey = process.env.FORGE_API_KEY ?? process.env.OPENAI_API_KEY ?? "";
1948
+ const apiKey = getAnvilEnv("ANVIL_API_KEY", "FORGE_API_KEY") ?? process.env.OPENAI_API_KEY ?? "";
1735
1949
  if (!apiKey)
1736
- throw new Error("Set FORGE_API_KEY (or OPENAI_API_KEY) before using --ai.");
1737
- const provider = new OpenAICompatibleProvider({ apiKey, model: process.env.FORGE_MODEL ?? "gpt-4.1-mini", baseUrl: process.env.FORGE_BASE_URL ?? "https://api.openai.com/v1" });
1950
+ throw new Error("Set ANVIL_API_KEY, FORGE_API_KEY, or OPENAI_API_KEY before using --ai.");
1951
+ const provider = new OpenAICompatibleProvider({ apiKey, model: getAnvilEnv("ANVIL_MODEL", "FORGE_MODEL", "gpt-4.1-mini"), baseUrl: getAnvilEnv("ANVIL_BASE_URL", "FORGE_BASE_URL", "https://api.openai.com/v1") });
1738
1952
  ai = await generateAIReview(report, provider);
1739
1953
  }
1740
1954
  const saved = await saveReview(target, report, ai);
@@ -1967,7 +2181,7 @@ async function main(args) {
1967
2181
  case "clean":
1968
2182
  case "fix": {
1969
2183
  if (rest.includes("--help")) {
1970
- process.stdout.write(`Usage: anvil ${command} [path] [--apply] [--fix] [--json]\n Clean farzi/junk files, downloads, logs, and auto-repair code issues.\n`);
2184
+ process.stdout.write(`Usage: anvil ${command} [path] [--apply] [--fix] [--json]\n Clean workspace junk clutter, downloads, logs, and auto-repair code issues.\n`);
1971
2185
  return;
1972
2186
  }
1973
2187
  const json = rest.includes("--json");