@kenkaiiii/ggcoder 5.62.0 → 5.64.0
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 +15 -15
- package/dist/app-sidecar.js +1 -1
- package/dist/app-sidecar.js.map +1 -1
- package/dist/cli.js +6 -6
- package/dist/cli.js.map +1 -1
- package/dist/core/agent-session.d.ts +13 -0
- package/dist/core/agent-session.js +44 -5
- package/dist/core/agent-session.js.map +1 -1
- package/dist/core/agents.js +14 -42
- package/dist/core/agents.js.map +1 -1
- package/dist/core/auth-providers.js +2 -2
- package/dist/core/auth-providers.js.map +1 -1
- package/dist/core/bundled-agents.js +9 -7
- package/dist/core/bundled-agents.js.map +1 -1
- package/dist/core/fast-apply-benchmark.d.ts +1 -1
- package/dist/core/fast-apply-benchmark.js +2 -2
- package/dist/core/fast-apply-benchmark.js.map +1 -1
- package/dist/core/frontmatter.d.ts +22 -0
- package/dist/core/frontmatter.js +90 -0
- package/dist/core/frontmatter.js.map +1 -0
- package/dist/core/regrounding.d.ts +7 -0
- package/dist/core/regrounding.js +27 -0
- package/dist/core/regrounding.js.map +1 -1
- package/dist/core/semantic-search-benchmark.d.ts +1 -1
- package/dist/core/semantic-search-benchmark.js +2 -2
- package/dist/core/semantic-search-benchmark.js.map +1 -1
- package/dist/core/skills.d.ts +1 -1
- package/dist/core/skills.js +5 -24
- package/dist/core/skills.js.map +1 -1
- package/dist/core/subagent-policy.js +1 -1
- package/dist/core/subagent-policy.js.map +1 -1
- package/dist/tools/edit.js +8 -1
- package/dist/tools/edit.js.map +1 -1
- package/dist/tools/redaction-guard.d.ts +10 -0
- package/dist/tools/redaction-guard.js +55 -0
- package/dist/tools/redaction-guard.js.map +1 -0
- package/dist/tools/steroids.d.ts +1 -1
- package/dist/tools/steroids.js +20 -4
- package/dist/tools/steroids.js.map +1 -1
- package/dist/tools/write.js +12 -5
- package/dist/tools/write.js.map +1 -1
- package/dist/ui/components/Footer.js +3 -4
- package/dist/ui/components/Footer.js.map +1 -1
- package/package.json +4 -4
package/README.md
CHANGED
|
@@ -69,8 +69,8 @@ Switch mid-conversation with `/model`. Not locked to anyone.
|
|
|
69
69
|
|
|
70
70
|
| Provider | Models | Auth |
|
|
71
71
|
| ----------------- | ------------------------------------------------------------------ | ---------------- |
|
|
72
|
-
| **Anthropic** | Claude Fable 5.1, Opus 5, Sonnet 5, Haiku 4.5
|
|
73
|
-
| **OpenAI** | GPT-6 Astra, GPT-
|
|
72
|
+
| **Anthropic** | Claude Fable 5.1, Opus 5.5, Sonnet 5, Haiku 4.5 | OAuth |
|
|
73
|
+
| **OpenAI** | GPT-6 Astra, GPT-6 Sol, GPT-6 Luna | OAuth |
|
|
74
74
|
| **Moonshot** | Kimi K3, Kimi K2.7 Code | OAuth or API key |
|
|
75
75
|
| **Z.AI (GLM)** | GLM-5.3, GLM-5.3-Flash (image) | API key |
|
|
76
76
|
| **MiniMax** | MiniMax M3 (image + video) | API key |
|
|
@@ -155,7 +155,7 @@ Plus the [Grep MCP](https://grep.dev) for searching across 1M+ public GitHub rep
|
|
|
155
155
|
|
|
156
156
|
`subagent` remains blocking. The async suite launches persistent NDJSON worker processes, so a parent can start up to four active child turns, keep working, steer them, and wait for any or all results. Up to eight idle workers remain available for follow-up; bounded snapshots retain the latest 20 agents.
|
|
157
157
|
|
|
158
|
-
Only GPT-
|
|
158
|
+
Only GPT-6 Astra/Sol at **Ultra** delegates proactively. Lower Astra/Sol levels use async agents only when the user or project/skill instructions request delegation; other models receive no proactive policy.
|
|
159
159
|
|
|
160
160
|
Children share the parent working directory, not isolated worktrees. Parallel writes must target disjoint files or subsystems. Async fan-out is one level deep, child output is bounded, idle workers reap after 10 minutes, and workers are not resumable after a CLI/app restart.
|
|
161
161
|
|
|
@@ -188,19 +188,19 @@ They get loaded into the system prompt automatically. The agent knows what it ca
|
|
|
188
188
|
|
|
189
189
|
Eleven ship built in, and route themselves when the work matches:
|
|
190
190
|
|
|
191
|
-
| Skill | Fires on
|
|
192
|
-
| ------------------ |
|
|
193
|
-
| `bulletproof` | Code an attacker will reach — auth, untrusted input, secrets, dependencies, CI/release, agent/MCP tool surfaces — and "is this safe to ship" reviews. Works on any target: web, API, CLI, desktop, mobile, embedded, contracts, ML.
|
|
194
|
-
| `clarify` | Requirements or a design genuinely unsettled — interrogating or stress-testing a plan before building, or a mid-build decision that materially changes the result.
|
|
195
|
-
| `code-review` | Reviewing written work — a diff, PR, or branch — on both what was asked for and how well it is built.
|
|
196
|
-
| `compliance-guard` | Legal exposure — personal data, payments, UGC, email/SMS, minors, or a licensed/regulated feature.
|
|
197
|
-
| `durable` | User data must not be lost — first database/table, migrations, backfills/imports, destructive operations, backups and recovery; any store (Postgres, MySQL, SQLite, Mongo, serverless).
|
|
198
|
-
| `evidence-led-ui` | Broad or design-sensitive UI work — new screens, redesigns, design systems, accessibility passes.
|
|
191
|
+
| Skill | Fires on |
|
|
192
|
+
| ------------------ | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
|
|
193
|
+
| `bulletproof` | Code an attacker will reach — auth, untrusted input, secrets, dependencies, CI/release, agent/MCP tool surfaces — and "is this safe to ship" reviews. Works on any target: web, API, CLI, desktop, mobile, embedded, contracts, ML. |
|
|
194
|
+
| `clarify` | Requirements or a design genuinely unsettled — interrogating or stress-testing a plan before building, or a mid-build decision that materially changes the result. |
|
|
195
|
+
| `code-review` | Reviewing written work — a diff, PR, or branch — on both what was asked for and how well it is built. |
|
|
196
|
+
| `compliance-guard` | Legal exposure — personal data, payments, UGC, email/SMS, minors, or a licensed/regulated feature. |
|
|
197
|
+
| `durable` | User data must not be lost — first database/table, migrations, backfills/imports, destructive operations, backups and recovery; any store (Postgres, MySQL, SQLite, Mongo, serverless). |
|
|
198
|
+
| `evidence-led-ui` | Broad or design-sensitive UI work — new screens, redesigns, design systems, accessibility passes. |
|
|
199
199
|
| `lean` | Speed and resource efficiency — slow loading/startup, jank, high CPU, memory leaks and hogging, zombie/orphan processes, bundle bloat, dead code/styles, Core Web Vitals; while building anything that should stay fast, or a perf pass on an existing project. Any stack: web, backend, Electron, Tauri, mobile, native, game, ML. |
|
|
200
|
-
| `refactoring` | Restructuring existing code without changing behavior — "refactor", "clean up", "reduce tech debt", "modernize" or migrate legacy code; test-guarded steps, revert-on-red, characterization tests for untested code.
|
|
201
|
-
| `root-cause` | A bug that resists the obvious fix, makes no sense, or keeps coming back — gated diagnosis from red repro to ranked hypotheses to regression test.
|
|
202
|
-
| `shared-language` | Fuzzy or drifting domain vocabulary, recurring naming decisions, and hard-to-reverse decisions worth recording (glossary + decision records).
|
|
203
|
-
| `tdd` | Test-driven development — red-green-refactor with pre-agreed seams, when the user asks for test-first work.
|
|
200
|
+
| `refactoring` | Restructuring existing code without changing behavior — "refactor", "clean up", "reduce tech debt", "modernize" or migrate legacy code; test-guarded steps, revert-on-red, characterization tests for untested code. |
|
|
201
|
+
| `root-cause` | A bug that resists the obvious fix, makes no sense, or keeps coming back — gated diagnosis from red repro to ranked hypotheses to regression test. |
|
|
202
|
+
| `shared-language` | Fuzzy or drifting domain vocabulary, recurring naming decisions, and hard-to-reverse decisions worth recording (glossary + decision records). |
|
|
203
|
+
| `tdd` | Test-driven development — red-green-refactor with pre-agreed seams, when the user asks for test-first work. |
|
|
204
204
|
|
|
205
205
|
---
|
|
206
206
|
|
package/dist/app-sidecar.js
CHANGED
|
@@ -512,7 +512,7 @@ async function runJsonModeIfRequested() {
|
|
|
512
512
|
await runJsonMode({
|
|
513
513
|
message: positionals[0] ?? "",
|
|
514
514
|
provider: (values.provider ?? "anthropic"),
|
|
515
|
-
model: values.model ?? "claude-opus-5",
|
|
515
|
+
model: values.model ?? "claude-opus-5-5",
|
|
516
516
|
cwd: process.cwd(),
|
|
517
517
|
systemPrompt: values["system-prompt"],
|
|
518
518
|
agentPrompt: values["agent-prompt"],
|