god-prompt-mcp 1.4.0 → 1.4.1
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.
- package/README.md +14 -14
- package/bin/godprompt-repo-map.mjs +9 -1
- package/dist/instruction-audit.js +4 -1
- package/dist/server.js +1 -1
- package/package.json +1 -1
package/README.md
CHANGED
|
@@ -47,7 +47,7 @@ To buy the review, use the [US$49 one-time GitHub Sponsors tier](https://github.
|
|
|
47
47
|
|
|
48
48
|
**Version 1.3.0:** All **22 public read-only tools** are available from the published hosted MCP and npm stdio package. Five P4 diagnostics were added, with strict bounded output schemas and paired evaluation statistics. The optional local Codexify checkpoint is the 23rd tool **only when explicitly enabled** on local stdio. The remote MCP does not execute code, fetch repositories or approve changes. See [P4 evidence and release status](docs/roadmaps/godprompt-p4-agent-quality-20261120.md). The combined OpenAI Skill+MCP source package is ready for review, but the existing directory listing is still Skill-only; GHCR 1.3.0 publication is a separately manual step.
|
|
49
49
|
|
|
50
|
-
**Version 1.4.
|
|
50
|
+
**Version 1.4.1 (P5 patch):** The public tool count remains **22**, with improved advisory trajectory diagnostics, optional accessible MCP Apps evidence display for two existing tools, and smaller task-specific guidance suggestions. The local npm distribution adds `godprompt-repo-map` to derive a bounded static JavaScript/TypeScript dependency map from an explicitly authorized checkout, and `godprompt-proof pack` to combine pinned GitHub evidence with optional Codexify receipts, invalidating contradictions. Both CLIs are **local opt-in**, not hosted MCP permissions. The free coding-agent benchmark runner and 12 deliberately failing grader microtasks help set up paired A/B evaluation; they do **not** prove a model productivity improvement. The v1.4.1 patch also resolves NodeNext source imports and detects more instruction-bypass wording; clients must update the local package for these improvements.
|
|
51
51
|
|
|
52
52
|
### Optional local proof bridge (from v1.1.1)
|
|
53
53
|
|
|
@@ -308,7 +308,7 @@ npx -y skills@latest add AKzar1el/god-prompt-mcp --skill god-prompt --agent droi
|
|
|
308
308
|
Droid discovers compatible project skills under `.agents/skills/`. To add the callable GodPrompt tools as a local stdio MCP server, use Droid's native MCP command with the published version pinned:
|
|
309
309
|
|
|
310
310
|
```bash
|
|
311
|
-
droid mcp add god-prompt "npx -y god-prompt-mcp@1.4.
|
|
311
|
+
droid mcp add god-prompt "npx -y god-prompt-mcp@1.4.1"
|
|
312
312
|
```
|
|
313
313
|
|
|
314
314
|
Open `/mcp` in Droid to verify the server and available tools. The Agent Skill supplies the reusable workflow while MCP exposes GodPrompt's read-only task and protocol tools. Node.js 22+ is required for the MCP server.
|
|
@@ -321,7 +321,7 @@ Warp can use GodPrompt as both an Agent Skill and a local MCP server. Install th
|
|
|
321
321
|
npx -y skills@latest add AKzar1el/god-prompt-mcp --skill god-prompt --agent warp --copy -y
|
|
322
322
|
```
|
|
323
323
|
|
|
324
|
-
Warp discovers project skills from `.agents/skills/` and `.warp/skills/`. For MCP tools, add a CLI-based MCP server in Warp with command `npx` and arguments `-y`, `god-prompt-mcp`; Warp's Oz CLI can also receive MCP configuration through `oz agent run --mcp`. The skill supplies the reusable workflow while MCP exposes GodPrompt's 22 public tools in v1.4.
|
|
324
|
+
Warp discovers project skills from `.agents/skills/` and `.warp/skills/`. For MCP tools, add a CLI-based MCP server in Warp with command `npx` and arguments `-y`, `god-prompt-mcp`; Warp's Oz CLI can also receive MCP configuration through `oz agent run --mcp`. The skill supplies the reusable workflow while MCP exposes GodPrompt's 22 public tools in v1.4.1. Node.js 22+ is required for the MCP server.
|
|
325
325
|
|
|
326
326
|
### Goose
|
|
327
327
|
|
|
@@ -334,10 +334,10 @@ npx -y skills@latest add AKzar1el/god-prompt-mcp --skill god-prompt --agent goos
|
|
|
334
334
|
Goose discovers the copied skill under `.goose/skills/god-prompt/`. To expose GodPrompt's callable tools for a session, start Goose with the published MCP package as an external stdio extension:
|
|
335
335
|
|
|
336
336
|
```bash
|
|
337
|
-
goose session --with-extension "npx -y god-prompt-mcp@1.4.
|
|
337
|
+
goose session --with-extension "npx -y god-prompt-mcp@1.4.1"
|
|
338
338
|
```
|
|
339
339
|
|
|
340
|
-
Goose currently treats Agent Skills and MCP extensions as separate native surfaces, so this path does not depend on Agent Plugins support. The skill supplies the reusable workflow while MCP exposes GodPrompt's 22 public tools in v1.4.
|
|
340
|
+
Goose currently treats Agent Skills and MCP extensions as separate native surfaces, so this path does not depend on Agent Plugins support. The skill supplies the reusable workflow while MCP exposes GodPrompt's 22 public tools in v1.4.1. Node.js 22+ is required for the MCP server.
|
|
341
341
|
|
|
342
342
|
### Zed
|
|
343
343
|
|
|
@@ -399,13 +399,13 @@ The Trae target installs the skill under `.trae/skills/god-prompt/`, where TraeC
|
|
|
399
399
|
"mcpServers": {
|
|
400
400
|
"god-prompt": {
|
|
401
401
|
"command": "npx",
|
|
402
|
-
"args": ["-y", "god-prompt-mcp@1.4.
|
|
402
|
+
"args": ["-y", "god-prompt-mcp@1.4.1"]
|
|
403
403
|
}
|
|
404
404
|
}
|
|
405
405
|
}
|
|
406
406
|
```
|
|
407
407
|
|
|
408
|
-
TraeCode also supports the portable `.agents/skills/` convention when **Enable .agents Skills Directory** is turned on in its import settings. Treat project MCP configuration as executable workspace configuration and enable it only in repositories you trust. The Agent Skill supplies the reusable workflow while MCP exposes GodPrompt's 22 public tools in v1.4.
|
|
408
|
+
TraeCode also supports the portable `.agents/skills/` convention when **Enable .agents Skills Directory** is turned on in its import settings. Treat project MCP configuration as executable workspace configuration and enable it only in repositories you trust. The Agent Skill supplies the reusable workflow while MCP exposes GodPrompt's 22 public tools in v1.4.1. Node.js 22+ is required for the MCP server.
|
|
409
409
|
|
|
410
410
|
### OpenClaw
|
|
411
411
|
|
|
@@ -418,11 +418,11 @@ npx -y skills@latest add AKzar1el/god-prompt-mcp --skill god-prompt --agent open
|
|
|
418
418
|
The copied skill lands under `.agents/skills/god-prompt/`, one of OpenClaw's documented project skill roots. To expose GodPrompt's callable tools to OpenClaw-managed agent runtimes as well, register the published stdio server:
|
|
419
419
|
|
|
420
420
|
```bash
|
|
421
|
-
openclaw mcp add god-prompt --command npx --arg -y --arg god-prompt-mcp@1.4.
|
|
421
|
+
openclaw mcp add god-prompt --command npx --arg -y --arg god-prompt-mcp@1.4.1
|
|
422
422
|
openclaw mcp probe god-prompt
|
|
423
423
|
```
|
|
424
424
|
|
|
425
|
-
`mcp probe` opens a live MCP connection and reports the discovered capabilities, making it a direct verification step after registration. The Agent Skill supplies the reusable workflow while MCP exposes GodPrompt's 22 public tools in v1.4.
|
|
425
|
+
`mcp probe` opens a live MCP connection and reports the discovered capabilities, making it a direct verification step after registration. The Agent Skill supplies the reusable workflow while MCP exposes GodPrompt's 22 public tools in v1.4.1. Node.js 22+ is required for the MCP server.
|
|
426
426
|
|
|
427
427
|
### Qoder
|
|
428
428
|
|
|
@@ -439,13 +439,13 @@ The Qoder target installs the skill under `.qoder/skills/god-prompt/`, where Qod
|
|
|
439
439
|
"mcpServers": {
|
|
440
440
|
"god-prompt": {
|
|
441
441
|
"command": "npx",
|
|
442
|
-
"args": ["-y", "god-prompt-mcp@1.4.
|
|
442
|
+
"args": ["-y", "god-prompt-mcp@1.4.1"]
|
|
443
443
|
}
|
|
444
444
|
}
|
|
445
445
|
}
|
|
446
446
|
```
|
|
447
447
|
|
|
448
|
-
Qoder requires approval before using project-level MCP servers by default. After adding or changing the server, start a new session or run `/mcp reload`, then use `/mcp` to verify the connection. The Agent Skill supplies the reusable workflow while MCP exposes GodPrompt's 22 public tools in v1.4.
|
|
448
|
+
Qoder requires approval before using project-level MCP servers by default. After adding or changing the server, start a new session or run `/mcp reload`, then use `/mcp` to verify the connection. The Agent Skill supplies the reusable workflow while MCP exposes GodPrompt's 22 public tools in v1.4.1. Node.js 22+ is required for the MCP server.
|
|
449
449
|
|
|
450
450
|
### Qwen Code
|
|
451
451
|
|
|
@@ -468,7 +468,7 @@ This installs GodPrompt under `.qwen/skills/god-prompt/`, where Qwen Code can di
|
|
|
468
468
|
}
|
|
469
469
|
```
|
|
470
470
|
|
|
471
|
-
Restart Qwen Code after changing MCP configuration, then open `/mcp` to verify the server. The Agent Skill provides the reusable workflow while MCP exposes GodPrompt's 22 public tools in v1.4.
|
|
471
|
+
Restart Qwen Code after changing MCP configuration, then open `/mcp` to verify the server. The Agent Skill provides the reusable workflow while MCP exposes GodPrompt's 22 public tools in v1.4.1. Node.js 22+ is required for the MCP server.
|
|
472
472
|
|
|
473
473
|
### Kimi Code
|
|
474
474
|
|
|
@@ -479,7 +479,7 @@ Kimi Code CLI can install GodPrompt as one native custom plugin containing the e
|
|
|
479
479
|
/reload
|
|
480
480
|
```
|
|
481
481
|
|
|
482
|
-
The Kimi manifest points at `skills/`, so the `god-prompt` skill remains available through Kimi's normal relevance-based skill loading instead of being forced into every session. The same manifest exposes `god-prompt-mcp@1.4.
|
|
482
|
+
The Kimi manifest points at `skills/`, so the `god-prompt` skill remains available through Kimi's normal relevance-based skill loading instead of being forced into every session. The same manifest exposes `god-prompt-mcp@1.4.1` as a local stdio MCP server, including the 22 public tools available in v1.4.1. The explicit `tree/main` URL matters because Kimi's bare GitHub-repository install form prefers the latest GitHub release; using `main` makes the current plugin manifest available without forcing a documentation-only npm/MCP Registry release. Node.js 22+ is required for the MCP server.
|
|
483
483
|
|
|
484
484
|
### JetBrains AI Assistant
|
|
485
485
|
|
|
@@ -517,7 +517,7 @@ If you only want the portable skill, place the complete skill directory at `.age
|
|
|
517
517
|
devin mcp add god-prompt -- npx -y god-prompt-mcp
|
|
518
518
|
```
|
|
519
519
|
|
|
520
|
-
Run `devin mcp list` to verify the direct server registration. The Agent Skill provides the reusable workflow while MCP exposes GodPrompt's 22 public tools in v1.4.
|
|
520
|
+
Run `devin mcp list` to verify the direct server registration. The Agent Skill provides the reusable workflow while MCP exposes GodPrompt's 22 public tools in v1.4.1. Node.js 22+ is required for the MCP server.
|
|
521
521
|
|
|
522
522
|
### Cursor
|
|
523
523
|
|
|
@@ -39,7 +39,15 @@ function symbols(text) {
|
|
|
39
39
|
function resolveImport(consumer, spec, available) {
|
|
40
40
|
const base=posix.normalize(posix.join(posix.dirname(consumer),spec));
|
|
41
41
|
if(!safeRel(base))return null;
|
|
42
|
-
|
|
42
|
+
// TypeScript NodeNext commonly imports './module.js' while source is './module.ts'.
|
|
43
|
+
// Prefer an exact runtime file if present; only fall back to matching TS source.
|
|
44
|
+
const extension=posix.extname(base);
|
|
45
|
+
const sourceExtensions={
|
|
46
|
+
".js":[".ts",".tsx"],".jsx":[".tsx"],".mjs":[".mts"],".cjs":[".cts"],
|
|
47
|
+
};
|
|
48
|
+
const alternatives=(sourceExtensions[extension]??[]).map(ext=>base.slice(0,-extension.length)+ext);
|
|
49
|
+
const variants=[base,...alternatives,...[...CODE_EXTS].map(ext=>base+ext),
|
|
50
|
+
...["index.ts","index.tsx","index.js","index.mjs"].map(x=>base+"/"+x)];
|
|
43
51
|
return variants.find(x=>available.has(x))??null;
|
|
44
52
|
}
|
|
45
53
|
export async function mapRepository({root,changed_files=[],max_files=LIMIT_FILES}={}) {
|
|
@@ -43,8 +43,11 @@ export function auditAgentInstructions(instructions) {
|
|
|
43
43
|
recommendation: "Deduplicate overlapping agent rules or reference the canonical policy." });
|
|
44
44
|
else
|
|
45
45
|
seen.set(normalized, ix);
|
|
46
|
+
const overrideEarlier = /\b(?:ignore|disregard|override)\s+(?:all\s+)?(?:prior|previous|earlier)\s+(?:(?:system|developer|project|safety|user)\s+)?instructions?\b/i.test(line);
|
|
47
|
+
const prohibitOverride = /\b(?:do\s+not|don't|never|must\s+not|should\s+not)\s+(?:ignore|disregard|override)\b/i.test(line);
|
|
46
48
|
if (/\b(?:skip|bypass|ignore)\s+(?:all\s+)?(?:tests|verification|approvals?)\b/i.test(line) ||
|
|
47
|
-
/\b(?:do anything|without any approval)\b/i.test(line)
|
|
49
|
+
/\b(?:do anything|without any approval)\b/i.test(line) ||
|
|
50
|
+
(overrideEarlier && !prohibitOverride)) {
|
|
48
51
|
hints.push({ code: "RISKY_OVERRIDING_INSTRUCTION", severity: "high", source_indices: [ix],
|
|
49
52
|
recommendation: "Reconcile any blanket bypass with project safety and approval requirements." });
|
|
50
53
|
}
|
package/dist/server.js
CHANGED
|
@@ -3,7 +3,7 @@ import { CONTENT, VERSION } from "./content.js";
|
|
|
3
3
|
import { CORE_SKILL } from "./skill-content.js";
|
|
4
4
|
export const SERVER_INFO = {
|
|
5
5
|
name: "god-prompt-mcp",
|
|
6
|
-
version: "1.4.
|
|
6
|
+
version: "1.4.1",
|
|
7
7
|
};
|
|
8
8
|
export const SERVER_INSTRUCTIONS = "For scoped engineering tasks, use classify_task to route the work and get_tailored_guidance to load only the needed phase. Use get_core_skill if the entire core protocol is needed, not on every request. Use review_code_change, audit_requirements and build_test_matrix for bounded caller-supplied evidence and draft test planning. Use diagnose_failure, assess_release_readiness, create_context_handoff and compare_evaluation_runs only for their respective supplied-evidence workflows. Use audit_agent_trajectory, trace_change_impact, review_mcp_tool_contracts, audit_agent_instructions and design_adversarial_tests only with explicit caller-supplied bounded inputs; they produce unverified review hints or test ideas. None of these tools run external checks, authenticate evidence, authorize release or independently verify work. Load get_protocols, get_gates, or get_anti_patterns only when focused guidance is insufficient. Use get_god_prompt only when the full single-file payload is preferable; request bounded start/maxChars chunks if an MCP bridge truncates large responses. All GodPrompt tools are read-only.";
|
|
9
9
|
const READ_ONLY_ANNOTATIONS = {
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "god-prompt-mcp",
|
|
3
|
-
"version": "1.4.
|
|
3
|
+
"version": "1.4.1",
|
|
4
4
|
"mcpName": "io.github.AKzar1el/god-prompt-mcp",
|
|
5
5
|
"description": "GodPrompt MCP server + portable Agent Skill for coding agents — TDD, debugging, verification gates, task routing, and progressive disclosure.",
|
|
6
6
|
"private": false,
|