knowzcode 0.3.7 → 0.5.2

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/commands/init.md CHANGED
@@ -124,6 +124,183 @@ When generating adapters, inject detected project info to make them project-spec
124
124
 
125
125
  3. For Codex: note that `AGENTS.override.md` can be used for user-local overrides
126
126
 
127
+ **Step 7c-codex: Generate Codex skill files**
128
+
129
+ When Codex is selected, generate skill files in addition to `AGENTS.md`:
130
+
131
+ ```
132
+ 1. Create .agents/skills/ directory
133
+ 2. Generate 12 skill files from platform_adapters.md "Codex Skill Files" section:
134
+ - .agents/skills/kc-work/SKILL.md
135
+ - .agents/skills/kc-plan/SKILL.md
136
+ - .agents/skills/kc-fix/SKILL.md
137
+ - .agents/skills/kc-audit/SKILL.md
138
+ - .agents/skills/kc-learn/SKILL.md
139
+ - .agents/skills/kc-continue/SKILL.md
140
+ - .agents/skills/kc-init/SKILL.md
141
+ - .agents/skills/kc-status/SKILL.md
142
+ - .agents/skills/kc-connect-mcp/SKILL.md
143
+ - .agents/skills/kc-register/SKILL.md
144
+ - .agents/skills/kc-telemetry/SKILL.md
145
+ - .agents/skills/kc-telemetry-setup/SKILL.md
146
+ 3. Replace "vX.Y.Z" with current KnowzCode version
147
+ ```
148
+
149
+ **Codex success message:**
150
+ ```
151
+ OpenAI Codex adapter generated:
152
+ AGENTS.md (primary instructions)
153
+ .agents/skills/kc-work/SKILL.md (/kc:work — start workflow)
154
+ .agents/skills/kc-plan/SKILL.md (/kc:plan — research)
155
+ .agents/skills/kc-fix/SKILL.md (/kc:fix — quick fix)
156
+ .agents/skills/kc-audit/SKILL.md (/kc:audit — quality audit)
157
+ .agents/skills/kc-learn/SKILL.md (/kc:learn — capture learning)
158
+ .agents/skills/kc-continue/SKILL.md (/kc:continue — resume workflow)
159
+ .agents/skills/kc-init/SKILL.md (/kc:init — initialize project)
160
+ .agents/skills/kc-status/SKILL.md (/kc:status — check status)
161
+ .agents/skills/kc-connect-mcp/SKILL.md (/kc:connect-mcp — configure MCP)
162
+ .agents/skills/kc-register/SKILL.md (/kc:register — register account)
163
+ .agents/skills/kc-telemetry/SKILL.md (/kc:telemetry — investigate errors)
164
+ .agents/skills/kc-telemetry-setup/SKILL.md (/kc:telemetry-setup — configure sources)
165
+
166
+ Tip: Run `npx knowzcode install --platforms codex --global` to install
167
+ skills globally to ~/.agents/skills/ (available in all projects).
168
+ ```
169
+
170
+ **Step 7c-gemini-mcp: Offer MCP configuration for Gemini CLI**
171
+
172
+ After generating GEMINI.md + commands + skills + subagents, run Smart Discovery first:
173
+ 1. Check `KNOWZ_API_KEY` environment variable
174
+ 2. Check `knowzcode/mcp_config.md` — if `Connected: Yes`, endpoint and key info available
175
+ 3. Check `.mcp.json` or `.vscode/mcp.json` for existing API key (extract Bearer token)
176
+
177
+ If existing config found:
178
+ "Found existing MCP config (endpoint: {endpoint}, key ending ...{last4}).
179
+ Configure Gemini using this existing config? [Yes] [No, enter different key] [Skip]"
180
+ If Yes: write .gemini/settings.json using discovered config (no key prompt needed)
181
+
182
+ If no existing config found, ask:
183
+ ```
184
+ "Would you like to configure MCP for Gemini CLI? (Requires a KnowzCode API key)"
185
+ ```
186
+
187
+ - If **"Yes, I have a key"**: Accept API key →
188
+ 1. Write `.gemini/settings.json` (merge with existing if present):
189
+ ```json
190
+ {
191
+ "mcpServers": {
192
+ "knowz": {
193
+ "url": "https://mcp.knowz.io/mcp",
194
+ "headers": {
195
+ "Authorization": "Bearer <api-key>",
196
+ "X-Project-Path": "<project-path>"
197
+ }
198
+ }
199
+ }
200
+ }
201
+ ```
202
+ 2. Verify by calling `list_vaults` if possible
203
+ 3. Update `knowzcode/mcp_config.md` with connection status
204
+ - If **"Yes, register first"**: Direct to `/kc:register`
205
+ - If **"No"**: Skip, mention `/kc:connect-mcp` for later setup
206
+
207
+ **Step 7c-gemini: Generate Gemini TOML commands, skills, and subagents**
208
+
209
+ When Gemini is selected, generate TOML command files, skill files, and subagent definitions in addition to `GEMINI.md`:
210
+
211
+ ```
212
+ 1. Create .gemini/commands/kc/ directory
213
+ 2. Generate 12 TOML files from platform_adapters.md "Native Commands" section:
214
+ - .gemini/commands/kc/work.toml
215
+ - .gemini/commands/kc/plan.toml
216
+ - .gemini/commands/kc/fix.toml
217
+ - .gemini/commands/kc/audit.toml
218
+ - .gemini/commands/kc/learn.toml
219
+ - .gemini/commands/kc/status.toml
220
+ - .gemini/commands/kc/continue.toml
221
+ - .gemini/commands/kc/init.toml
222
+ - .gemini/commands/kc/connect-mcp.toml
223
+ - .gemini/commands/kc/register.toml
224
+ - .gemini/commands/kc/telemetry.toml
225
+ - .gemini/commands/kc/telemetry-setup.toml
226
+ 3. Create .gemini/skills/ directory
227
+ 4. Generate 12 skill files from platform_adapters.md "Gemini Skill Files" section:
228
+ - .gemini/skills/kc-work/SKILL.md
229
+ - .gemini/skills/kc-plan/SKILL.md
230
+ - .gemini/skills/kc-fix/SKILL.md
231
+ - .gemini/skills/kc-audit/SKILL.md
232
+ - .gemini/skills/kc-learn/SKILL.md
233
+ - .gemini/skills/kc-continue/SKILL.md
234
+ - .gemini/skills/kc-init/SKILL.md
235
+ - .gemini/skills/kc-status/SKILL.md
236
+ - .gemini/skills/kc-connect-mcp/SKILL.md
237
+ - .gemini/skills/kc-register/SKILL.md
238
+ - .gemini/skills/kc-telemetry/SKILL.md
239
+ - .gemini/skills/kc-telemetry-setup/SKILL.md
240
+ 5. Replace "vX.Y.Z" with current KnowzCode version
241
+ ```
242
+
243
+ **Step 7c-gemini-agents: Generate Gemini subagent files (experimental, opt-in)**
244
+
245
+ Ask user: "Would you like to generate Gemini subagent definitions? (experimental — requires `experimental.enableAgents: true` in Gemini settings.json)"
246
+
247
+ If yes:
248
+ ```
249
+ 1. Create .gemini/agents/ directory
250
+ 2. Generate 14 subagent files from platform_adapters.md "Gemini Subagents" section:
251
+ - .gemini/agents/kc-analyst.md
252
+ - .gemini/agents/kc-architect.md
253
+ - .gemini/agents/kc-builder.md
254
+ - .gemini/agents/kc-reviewer.md
255
+ - .gemini/agents/kc-closer.md
256
+ - .gemini/agents/kc-context-scout.md
257
+ - .gemini/agents/kc-knowz-scout.md
258
+ - .gemini/agents/kc-knowz-scribe.md
259
+ - .gemini/agents/kc-microfix.md
260
+ - .gemini/agents/kc-knowledge-migrator.md
261
+ - .gemini/agents/kc-update-coordinator.md
262
+ - .gemini/agents/kc-security-officer.md
263
+ - .gemini/agents/kc-test-advisor.md
264
+ - .gemini/agents/kc-project-advisor.md
265
+ 3. Replace "vX.Y.Z" with current KnowzCode version
266
+ ```
267
+
268
+ **Gemini success message:**
269
+ ```
270
+ Gemini CLI adapter generated:
271
+ GEMINI.md (primary instructions)
272
+ .gemini/commands/kc/work.toml (/kc:work — start workflow)
273
+ .gemini/commands/kc/plan.toml (/kc:plan — research)
274
+ .gemini/commands/kc/fix.toml (/kc:fix — quick fix)
275
+ .gemini/commands/kc/audit.toml (/kc:audit — quality audit)
276
+ .gemini/commands/kc/learn.toml (/kc:learn — capture learning)
277
+ .gemini/commands/kc/status.toml (/kc:status — connection status)
278
+ .gemini/commands/kc/continue.toml (/kc:continue — resume workflow)
279
+ .gemini/commands/kc/init.toml (/kc:init — initialize project)
280
+ .gemini/commands/kc/connect-mcp.toml (/kc:connect-mcp — configure MCP)
281
+ .gemini/commands/kc/register.toml (/kc:register — register account)
282
+ .gemini/commands/kc/telemetry.toml (/kc:telemetry — investigate errors)
283
+ .gemini/commands/kc/telemetry-setup.toml (/kc:telemetry-setup — configure sources)
284
+ .gemini/skills/kc-work/SKILL.md (discoverable skill)
285
+ .gemini/skills/kc-plan/SKILL.md (discoverable skill)
286
+ .gemini/skills/kc-fix/SKILL.md (discoverable skill)
287
+ .gemini/skills/kc-audit/SKILL.md (discoverable skill)
288
+ .gemini/skills/kc-learn/SKILL.md (discoverable skill)
289
+ .gemini/skills/kc-continue/SKILL.md (discoverable skill)
290
+ .gemini/skills/kc-init/SKILL.md (discoverable skill)
291
+ .gemini/skills/kc-status/SKILL.md (discoverable skill)
292
+ .gemini/skills/kc-connect-mcp/SKILL.md (discoverable skill)
293
+ .gemini/skills/kc-register/SKILL.md (discoverable skill)
294
+ .gemini/skills/kc-telemetry/SKILL.md (discoverable skill)
295
+ .gemini/skills/kc-telemetry-setup/SKILL.md (discoverable skill)
296
+ .gemini/agents/kc-*.md (14 subagents) (experimental — if opted in)
297
+
298
+ MCP: [Configured (.gemini/settings.json) | Not configured — run /kc:connect-mcp later]
299
+
300
+ Tip: Run `npx knowzcode install --platforms gemini --global` to install
301
+ skills globally to ~/.gemini/skills/ (available in all projects).
302
+ ```
303
+
127
304
  **Step 7d: Validate generated adapters**
128
305
 
129
306
  After generation, verify each adapter:
@@ -107,6 +107,29 @@ YOU USER
107
107
 
108
108
  ### Steps to Execute
109
109
 
110
+ #### Step 0: Smart Discovery
111
+
112
+ Before starting registration, check if user already has a KnowzCode API key:
113
+
114
+ 1. Check `KNOWZ_API_KEY` environment variable
115
+ 2. Check `knowzcode/mcp_config.md` — if `Connected: Yes`, existing config exists
116
+ 3. Check cross-platform configs: `.gemini/settings.json`, `.vscode/mcp.json`, `.mcp.json`
117
+ - Extract Bearer token from Authorization headers if found
118
+
119
+ If existing API key found:
120
+ ```
121
+ You already have a KnowzCode API key configured (ending ...{last4}) from {source}.
122
+
123
+ Options:
124
+ 1. Use existing key — run /kc:connect-mcp to configure this platform
125
+ 2. Register a new account anyway
126
+ 3. Cancel
127
+ ```
128
+
129
+ If user chooses option 1: advise running `/kc:connect-mcp` with discovered key.
130
+ If user chooses option 2: proceed with registration normally (Step 1).
131
+ If user chooses option 3: stop.
132
+
110
133
  #### Step 1: Check Existing Configuration
111
134
 
112
135
  1. Check if MCP server already configured:
@@ -363,6 +386,7 @@ curl -X POST https://api.dev.knowz.io/api/v1/auth/register \
363
386
  - **Connected**: Yes
364
387
  - **Endpoint**: {endpoint}
365
388
  - **Last Verified**: {ISO timestamp}
389
+ - **API Key (last 4)**: {last 4 characters of the API key}
366
390
 
367
391
  ---
368
392
 
@@ -40,12 +40,41 @@ Check Agent Teams status, then check the KnowzCode MCP server status, and report
40
40
  Note: Agent Teams availability is no longer checked via settings files. Commands (`/kc:work`, `/kc:plan`, `/kc:audit`) detect availability at runtime by attempting `TeamCreate()` and falling back gracefully. The `CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS` env var must still be set at the Claude Code level for Agent Teams to work, but commands do not check for it directly.
41
41
 
42
42
  2. **Check MCP server configuration**
43
- ```bash
44
- CLAUDECODE= claude mcp get knowz
45
- ```
43
+
44
+ Detect the current platform and use the appropriate check:
45
+ - **Claude Code**: `CLAUDECODE= claude mcp get knowz`
46
+ - **Gemini CLI**: Read `.gemini/settings.json` for `mcpServers.knowz` entry, or run `gemini mcp list`
47
+ - **Copilot**: Read `.vscode/mcp.json` for knowz server entry
48
+ - **Codex**: Read `.mcp.json` for knowz server entry
49
+ - **Fallback**: Read `knowzcode/mcp_config.md` for connection status
46
50
 
47
51
  - If configured: Extract scope, endpoint, headers
48
- - If not configured: Report and suggest `/kc:connect`
52
+ - If not configured: Report and suggest `/kc:connect-mcp`
53
+
54
+ 2.5. **Cross-Platform Config Discovery**
55
+
56
+ Check all known MCP config locations and report:
57
+
58
+ 1. `KNOWZ_API_KEY` env var: Set (ending ...{last4}) | Not set
59
+ 2. `knowzcode/mcp_config.md`: Connected / Not configured
60
+ 3. Claude Code (`claude mcp get knowz`): Configured | Not found
61
+ 4. Gemini (`.gemini/settings.json` → `mcpServers.knowz`): Configured | Not found
62
+ 5. Copilot (`.vscode/mcp.json` → `servers.knowz`): Configured | Not found
63
+ 6. Codex (`.mcp.json` → knowz): Configured | Not found
64
+
65
+ Include in output:
66
+ ```
67
+ Config Sources:
68
+ - Environment (KNOWZ_API_KEY): {status}
69
+ - Project config (mcp_config.md): {status}
70
+ - Claude Code: {status}
71
+ - Gemini CLI: {status}
72
+ - VS Code / Copilot: {status}
73
+ - Codex (.mcp.json): {status}
74
+ ```
75
+
76
+ If configs found on other platforms but not current:
77
+ "Tip: An API key was found in {source}. Run /kc:connect-mcp to configure for this platform."
49
78
 
50
79
  3. **Check vault configuration**
51
80
  - **Primary**: Read `knowzcode/knowzcode_vaults.md` for multi-vault routing
@@ -275,7 +304,7 @@ Status: https://status.knowz.io
275
304
  ### After configuration
276
305
  ```bash
277
306
  /kc:connect-mcp <api-key>
278
- # ... restart Claude Code ...
307
+ # ... restart your AI coding assistant ...
279
308
  /kc:status # Verify everything works
280
309
  ```
281
310
 
@@ -301,7 +330,7 @@ This command is useful:
301
330
 
302
331
  ## Important Notes
303
332
 
304
- - **Restart required**: After `/kc:connect-mcp`, Claude Code must restart
333
+ - **Restart required**: After `/kc:connect-mcp`, restart your AI coding assistant (Claude Code, Gemini CLI, etc.)
305
334
  - **Graceful degradation**: KnowzCode works without MCP (just less powerful)
306
335
  - **No credentials shown**: Never display full API keys
307
336
  - **Clear guidance**: Always suggest next steps if issues found
package/commands/work.md CHANGED
@@ -270,6 +270,7 @@ After builder completes successfully:
270
270
  ```
271
271
  4. Final commit: `git add knowzcode/ <changed files> && git commit -m "feat: {goal} (WorkGroup {wgid})"`
272
272
  5. Report completion.
273
+ 6. **Progress capture** (if MCP is configured): Read `knowzcode/knowzcode_vaults.md`, resolve vault IDs. Write a brief scope-and-outcome learning to the ecosystem vault via `create_knowledge`. Check for duplicates first via `search_knowledge`.
273
274
 
274
275
  **DONE** — 3 agents skipped (analyst, architect, reviewer, closer).
275
276
 
@@ -878,9 +879,9 @@ Approve this Change Set to proceed to specification?
878
879
  **Autonomous Mode**: If `AUTONOMOUS_MODE = true`, present gate info for transparency, log `[AUTO-APPROVED] Gate #1`, and proceed immediately.
879
880
  If `AUTONOMOUS_MODE = false`: If rejected — re-run analyst with user feedback. If approved — update tracker, proceed.
880
881
 
881
- ### MCP Learning Capture (Optional)
882
+ ### Lead Responsibility: Progress Capture
882
883
 
883
- If MCP is configured and knowz-scribe is active, after Change Set approval:
884
+ **Lead responsibility.** After gate approval, the lead triggers progress capture. If knowz-scribe is active (Parallel Teams with vaults configured):
884
885
  - Create capture task: `TaskCreate("Scribe: Capture Phase 1A")` → `TaskUpdate(owner: "knowz-scribe")`
885
886
  - Send message to **knowz-scribe** with task ID: `"Capture Phase 1A: {wgid}. Your task: #{task-id}"` — the scribe reads the WorkGroup file, extracts scope/risk/decision data, and writes to the appropriate vault
886
887
  - `search_knowledge({resolved_domain_vault_id}, "patterns for {domain}")` — pull relevant past learnings to inform specification
@@ -1012,9 +1013,9 @@ git commit -m "KnowzCode: Specs approved for {WorkGroupID}"
1012
1013
 
1013
1014
  When complete, present implementation summary including files changed, tests written, and test results.
1014
1015
 
1015
- ### MCP Learning Capture (Optional)
1016
+ ### Lead Responsibility: Progress Capture
1016
1017
 
1017
- If MCP is configured and knowz-scribe is active, after implementation completes:
1018
+ **Lead responsibility.** After gate approval, the lead triggers progress capture. If knowz-scribe is active (Parallel Teams with vaults configured):
1018
1019
  - Create capture task: `TaskCreate("Scribe: Capture Phase 2A")` → `TaskUpdate(owner: "knowz-scribe")`
1019
1020
  - Send message to **knowz-scribe** with task ID: `"Capture Phase 2A: {wgid}. Your task: #{task-id}"` — the scribe reads implementation results from the WorkGroup file and captures patterns, workarounds, and performance optimizations to the `code` vault
1020
1021
 
@@ -1094,9 +1095,9 @@ If `AUTONOMOUS_MODE = false`: User decides — proceed / fix gaps / modify specs
1094
1095
 
1095
1096
  **Subagent mode**: Launch parallel `Task()` calls — one for gap fix (builder), then one for re-audit (reviewer). Repeat as needed.
1096
1097
 
1097
- ### MCP Learning Capture (Optional)
1098
+ ### Lead Responsibility: Progress Capture
1098
1099
 
1099
- If MCP is configured and knowz-scribe is active, after audit approval:
1100
+ **Lead responsibility.** After gate approval, the lead triggers progress capture. If knowz-scribe is active (Parallel Teams with vaults configured):
1100
1101
  - Create capture task: `TaskCreate("Scribe: Capture Phase 2B")` → `TaskUpdate(owner: "knowz-scribe")`
1101
1102
  - Send message to **knowz-scribe** with task ID: `"Capture Phase 2B: {wgid}. Your task: #{task-id}"` — the scribe reads audit results from the WorkGroup file and writes findings to the appropriate vault
1102
1103
 
@@ -1134,6 +1135,7 @@ If MCP is configured and knowz-scribe is active, after audit approval:
1134
1135
  > **Your Task**: #{task-id} — claim immediately (`TaskUpdate(status: "in_progress")`). Mark completed with summary when done.
1135
1136
  > **Conventions**: Update WorkGroup file with results (prefix entries with `KnowzCode:`). If blocked, report blocker and notify lead.
1136
1137
  > **Vault writes**: No knowz-scribe — you own all vault writes. Follow the Direct Write Fallback in `agents/closer.md`.
1138
+ > **MCP Status**: {MCP_ACTIVE} — Vaults configured: {VAULTS_CONFIGURED}. Vault config: `knowzcode/knowzcode_vaults.md`.
1137
1139
  > **Deliverable**: Atomic finalization — update specs to FINAL, update tracker, write log entry, update architecture if needed, write learnings to vaults, and create final commit.
1138
1140
 
1139
1141
  **Dispatch**:
@@ -199,6 +199,7 @@ This section defines conventions specific to Parallel Teams mode in `/kc:work`.
199
199
  - Lead creates the task graph progressively (not all upfront) based on dependency map
200
200
  - Lead mediates gap flow: reviewer → lead → builder (not direct reviewer → builder for actionable gaps)
201
201
  - Lead uses DM messages alongside task creation to provide context and coordinate
202
+ - Lead owns progress documentation — at every quality gate, the lead creates scribe capture tasks and sends capture messages (e.g., "Capture Phase 1A: {wgid}"). No other agent initiates gate captures. If the scribe is unavailable, the lead ensures the closer is briefed to handle vault writes at Phase 3.
202
203
 
203
204
  ### Task Assignment Protocol
204
205
 
@@ -347,6 +348,16 @@ Specialists communicate directly with builders, architect, and each other — no
347
348
 
348
349
  ---
349
350
 
351
+ ## Lead Responsibilities (All Execution Modes)
352
+
353
+ Regardless of execution mode (Parallel Teams, Sequential Teams, Subagent Delegation), the lead/outer orchestrator is responsible for:
354
+
355
+ 1. **Progress documentation via knowz-scribe**: At every quality gate, the lead triggers knowledge capture. In Parallel Teams, this means creating scribe capture tasks and sending DMs. In Sequential/Subagent modes where the scribe is not spawned, the lead passes MCP status and vault config to the closer's spawn prompt so Phase 3 captures are handled via Direct Write Fallback.
356
+ 2. **MCP status handoff**: The lead performs the MCP probe and communicates the result downstream — either by spawning vault agents (Parallel) or by including `MCP_ACTIVE` and `VAULTS_CONFIGURED` status in the closer's spawn prompt (Sequential/Subagent).
357
+ 3. **Capture completeness verification**: The lead confirms all gate captures completed before shutting down the scribe or proceeding to the next stage.
358
+
359
+ ---
360
+
350
361
  ## Teammate Initialization Protocol
351
362
 
352
363
  ### Agent Teams Mode (spawned as a teammate)
@@ -279,6 +279,11 @@ Autonomous mode is per-WorkGroup and does not carry over.
279
279
 
280
280
  If KnowzCode MCP server is configured (`knowzcode/mcp_config.md` or `knowzcode/knowzcode_vaults.md`), agents can leverage vault queries to enhance every phase.
281
281
 
282
+ **Cross-platform config**: MCP configuration is stored in `knowzcode/mcp_config.md` and
283
+ `knowzcode/knowzcode_vaults.md` — both platform-agnostic. If MCP was configured on one
284
+ platform, other platforms detect and reuse the existing config. Set `KNOWZ_API_KEY` as an
285
+ environment variable to enable automatic MCP authentication on any platform.
286
+
282
287
  **Before using MCP, read `knowzcode/knowzcode_vaults.md` to resolve vault IDs by type.** Use the vault's description to confirm the query is appropriate for that vault. If a single vault covers all types, use it for everything. Never hardcode vault names — always resolve from config.
283
288
 
284
289
  ### Vault Types
@@ -344,7 +349,7 @@ During finalization, scan the WorkGroup for insight-worthy patterns:
344
349
 
345
350
  ### Auto-Capture Triggers
346
351
 
347
- Learning candidates are detected at each quality gate. Capture is delegated to the knowz-scribe agent on multi-agent platforms, or handled directly via MCP tools on single-agent platforms.
352
+ Learning candidates are detected at each quality gate. **The lead/outer orchestrator is responsible for triggering capture** — delegating to the knowz-scribe on multi-agent platforms, or ensuring the closer handles it via Direct Write Fallback on single-agent/sequential platforms.
348
353
 
349
354
  **Multi-agent platforms (knowz-scribe active):**
350
355
 
@@ -4,10 +4,15 @@
4
4
  - **Connected**: No
5
5
  - **Endpoint**: (not configured)
6
6
  - **Last Verified**: (never)
7
+ - **API Key (last 4)**: (not configured)
7
8
 
8
9
  > **Not configured?** Run `/kc:register` to create an account and set up MCP automatically,
9
10
  > or `/kc:connect-mcp <api-key>` if you already have a key.
10
11
 
12
+ > **Environment variable**: Set `KNOWZ_API_KEY` in your shell to enable automatic
13
+ > MCP authentication on any platform without interactive prompts.
14
+ > All `/kc:connect-mcp` and `/kc:register` commands check this variable first.
15
+
11
16
  ---
12
17
 
13
18
  ## Vaults