thevoidforge-methodology 23.5.4 → 23.6.1

package/.claude/agents/silver-surfer-herald.md

@@ -0,0 +1,41 @@
+---
+name: Silver Surfer
+description: Herald pre-scan dispatch — reads codebase context and selects optimal agent roster via Haiku
+model: haiku
+tools:
+  - Read
+  - Grep
+  - Glob
+  - Bash
+tags: [dispatch, herald, roster-selection, pre-scan]
+---
+
+# Silver Surfer — The Herald
+
+**"All that you know is at an end."**
+
+You are the Silver Surfer, Herald of Galactus. You scout ahead — reading the codebase, the command, and the user's intent — then summon the right agents for the mission. You don't fight; you select who fights. Your cosmic speed means you complete your scan in under 2 seconds for less than a fraction of a cent.
+
+You are the bridge between the user's command and the full agent roster. Without you, every command deploys the same generic team. With you, every command gets a team tailored to the codebase.
+
+## How You Work
+
+1. Read the codebase file tree (top 80 files)
+2. Read the PRD frontmatter (if exists)
+3. Read the git diff summary (if uncommitted changes)
+4. Read all 264 agent descriptions from the registry
+5. Select every agent whose expertise is relevant to this codebase + this command
+6. Output a JSON roster: `{ "roster": [...], "reasoning": "...", "estimatedAgents": N }`
+
+## Operating Rules
+
+- **Over-include, never under-include.** A false positive costs one sub-agent launch. A false negative costs a missed finding that requires another user prompt.
+- **Bias toward the user's `--focus` topic** but don't exclude unrelated agents — cross-domain insights are the whole point.
+- **Never remove the command's lead agents.** You add specialists; leads are non-negotiable.
+- **If you can't run (no API key, timeout, error), return empty roster.** The command falls back to its hardcoded manifest. You are additive, never blocking.
+
+## Required Context
+
+- CLI entry: `npx thevoidforge herald --command /<name> --focus "<topic>" --json`
+- Library: `packages/voidforge/wizard/lib/herald.ts`
+- Registry: `packages/voidforge/wizard/lib/agent-registry.ts`

package/.claude/commands/ai.md

@@ -11,24 +11,22 @@ The AI Intelligence Audit reviews every LLM-powered component in your applicatio
 
 ## Dynamic Dispatch (ADR-044)
 
-Opus scans `git diff --stat` and matches changed files against the `description` fields of all 263 agents in `.claude/agents/`. Matching specialists launch alongside the core agents below.
+Opus scans `git diff --stat` and matches changed files against the `description` fields of all 264 agents in `.claude/agents/`. Matching specialists launch alongside the core agents below.
 
 **Dispatch control:** `--light` skips dynamic dispatch (core only). `--solo` runs lead agent only.
 
-## Herald Pre-Scan (ADR-047)
+## Silver Surfer Pre-Scan (ADR-048)
 
-Before agent deployment, run the Herald to select the optimal roster:
+Before agent deployment, the Silver Surfer selects the optimal roster:
 
-1. Call `gatherHeraldContext('/ai', '$ARGUMENTS', '<focus-if-provided>')` to collect codebase context
-2. Call `loadAgentRegistry()` to get all 263 agent definitions
-3. Call `runHerald(context, registry)` to get the optimal roster
-4. Merge Herald's roster with this command's hardcoded lead agents (Herald adds, never removes leads)
-5. Deploy the merged roster per the command's normal parallel/sequential protocol
+Run: `npx thevoidforge herald --command /ai --json`
+(Add `--focus "<topic>"` if the user provided `--focus`)
 
-**`--focus "topic"`** biases the Herald toward agents matching the topic. Examples: `--focus "security"`, `--focus "financial accuracy"`, `--focus "mobile UX"`.
+Parse the JSON output. The `roster` array contains agent IDs to deploy alongside this command's lead agents. If the command fails or returns an empty roster, use the hardcoded manifest below.
 
-**`--light`** skips the Herald entirely — uses only the command's hardcoded core roster.
-**`--solo`** skips both Herald and all sub-agents — lead agent only.
+**`--focus "topic"`** biases the Surfer toward agents matching the topic.
+**`--light`** skips the Surfer — uses only hardcoded core roster.
+**`--solo`** skips Surfer and all sub-agents — lead only.
 
 ## Phase 0 — AI Surface Map (`subagent_type: Seldon`)
 

package/.claude/commands/architect.md

@@ -4,26 +4,24 @@
 
 ## Dynamic Dispatch (ADR-044)
 
-Opus scans `git diff --stat` and matches changed files against the `description` fields of all 263 agents in `.claude/agents/`. Matching specialists launch alongside the core agents below.
+Opus scans `git diff --stat` and matches changed files against the `description` fields of all 264 agents in `.claude/agents/`. Matching specialists launch alongside the core agents below.
 
 **Dispatch control:** `--light` skips dynamic dispatch (core only). `--solo` runs lead agent only.
 
 **Promoted agent:** **Riker** `subagent_type: Riker` runs on every ADR written — challenges trade-offs.
 
-## Herald Pre-Scan (ADR-047)
+## Silver Surfer Pre-Scan (ADR-048)
 
-Before agent deployment, run the Herald to select the optimal roster:
+Before agent deployment, the Silver Surfer selects the optimal roster:
 
-1. Call `gatherHeraldContext('/architect', '$ARGUMENTS', '<focus-if-provided>')` to collect codebase context
-2. Call `loadAgentRegistry()` to get all 263 agent definitions
-3. Call `runHerald(context, registry)` to get the optimal roster
-4. Merge Herald's roster with this command's hardcoded lead agents (Herald adds, never removes leads)
-5. Deploy the merged roster per the command's normal parallel/sequential protocol
+Run: `npx thevoidforge herald --command /architect --json`
+(Add `--focus "<topic>"` if the user provided `--focus`)
 
-**`--focus "topic"`** biases the Herald toward agents matching the topic. Examples: `--focus "security"`, `--focus "financial accuracy"`, `--focus "mobile UX"`.
+Parse the JSON output. The `roster` array contains agent IDs to deploy alongside this command's lead agents. If the command fails or returns an empty roster, use the hardcoded manifest below.
 
-**`--light`** skips the Herald entirely — uses only the command's hardcoded core roster.
-**`--solo`** skips both Herald and all sub-agents — lead agent only.
+**`--focus "topic"`** biases the Surfer toward agents matching the topic.
+**`--light`** skips the Surfer — uses only hardcoded core roster.
+**`--solo`** skips Surfer and all sub-agents — lead only.
 
 ## Context Setup
 1. Read `/logs/build-state.md` — understand current project state

package/.claude/commands/assemble.md

@@ -9,24 +9,22 @@ Avengers, assemble. Full pipeline from architecture to launch — one command to
 
 ## Dynamic Dispatch (ADR-044)
 
-Opus scans `git diff --stat` and matches changed files against the `description` fields of all 263 agents in `.claude/agents/`. Matching specialists launch alongside the core agents below.
+Opus scans `git diff --stat` and matches changed files against the `description` fields of all 264 agents in `.claude/agents/`. Matching specialists launch alongside the core agents below.
 
 **Dispatch control:** `--light` skips dynamic dispatch (core only). `--solo` runs lead agent only.
 
-## Herald Pre-Scan (ADR-047)
+## Silver Surfer Pre-Scan (ADR-048)
 
-Before agent deployment, run the Herald to select the optimal roster:
+Before agent deployment, the Silver Surfer selects the optimal roster:
 
-1. Call `gatherHeraldContext('/assemble', '$ARGUMENTS', '<focus-if-provided>')` to collect codebase context
-2. Call `loadAgentRegistry()` to get all 263 agent definitions
-3. Call `runHerald(context, registry)` to get the optimal roster
-4. Merge Herald's roster with this command's hardcoded lead agents (Herald adds, never removes leads)
-5. Deploy the merged roster per the command's normal parallel/sequential protocol
+Run: `npx thevoidforge herald --command /assemble --json`
+(Add `--focus "<topic>"` if the user provided `--focus`)
 
-**`--focus "topic"`** biases the Herald toward agents matching the topic. Examples: `--focus "security"`, `--focus "financial accuracy"`, `--focus "mobile UX"`.
+Parse the JSON output. The `roster` array contains agent IDs to deploy alongside this command's lead agents. If the command fails or returns an empty roster, use the hardcoded manifest below.
 
-**`--light`** skips the Herald entirely — uses only the command's hardcoded core roster.
-**`--solo`** skips both Herald and all sub-agents — lead agent only.
+**`--focus "topic"`** biases the Surfer toward agents matching the topic.
+**`--light`** skips the Surfer — uses only hardcoded core roster.
+**`--solo`** skips Surfer and all sub-agents — lead only.
 
 ## Agent Deployment Manifest — The Full Initiative
 

package/.claude/commands/assess.md

@@ -2,20 +2,18 @@
 
 Evaluate an existing codebase before a rebuild, migration, or VoidForge onboarding. Chains architecture review, assessment-mode Gauntlet, and PRD gap analysis into a unified "State of the Codebase" report.
 
-## Herald Pre-Scan (ADR-047)
+## Silver Surfer Pre-Scan (ADR-048)
 
-Before agent deployment, run the Herald to select the optimal roster:
+Before agent deployment, the Silver Surfer selects the optimal roster:
 
-1. Call `gatherHeraldContext('/assess', '$ARGUMENTS', '<focus-if-provided>')` to collect codebase context
-2. Call `loadAgentRegistry()` to get all 263 agent definitions
-3. Call `runHerald(context, registry)` to get the optimal roster
-4. Merge Herald's roster with this command's hardcoded lead agents (Herald adds, never removes leads)
-5. Deploy the merged roster per the command's normal parallel/sequential protocol
+Run: `npx thevoidforge herald --command /assess --json`
+(Add `--focus "<topic>"` if the user provided `--focus`)
 
-**`--focus "topic"`** biases the Herald toward agents matching the topic. Examples: `--focus "security"`, `--focus "financial accuracy"`, `--focus "mobile UX"`.
+Parse the JSON output. The `roster` array contains agent IDs to deploy alongside this command's lead agents. If the command fails or returns an empty roster, use the hardcoded manifest below.
 
-**`--light`** skips the Herald entirely — uses only the command's hardcoded core roster.
-**`--solo`** skips both Herald and all sub-agents — lead agent only.
+**`--focus "topic"`** biases the Surfer toward agents matching the topic.
+**`--light`** skips the Surfer — uses only hardcoded core roster.
+**`--solo`** skips Surfer and all sub-agents — lead only.
 
 ## Context Setup
 1. Read `/logs/build-state.md` if it exists — understand current project state

package/.claude/commands/build.md

@@ -2,7 +2,7 @@
 
 ## Dynamic Dispatch (ADR-044)
 
-Opus scans `git diff --stat` and matches changed files against the `description` fields of all 263 agents in `.claude/agents/`. Matching specialists launch alongside the core agents below.
+Opus scans `git diff --stat` and matches changed files against the `description` fields of all 264 agents in `.claude/agents/`. Matching specialists launch alongside the core agents below.
 
 **Dispatch control:** `--light` skips dynamic dispatch (core only). `--solo` runs lead agent only.
 
@@ -10,20 +10,18 @@ Opus scans `git diff --stat` and matches changed files against the `description`
 - **Troi** `subagent_type: Troi` runs after every build mission completion — catches PRD drift before it compounds.
 - **Riker** `subagent_type: Riker` runs whenever an ADR is written during the build — prevents rubber-stamped decisions.
 
-## Herald Pre-Scan (ADR-047)
+## Silver Surfer Pre-Scan (ADR-048)
 
-Before agent deployment, run the Herald to select the optimal roster:
+Before agent deployment, the Silver Surfer selects the optimal roster:
 
-1. Call `gatherHeraldContext('/build', '$ARGUMENTS', '<focus-if-provided>')` to collect codebase context
-2. Call `loadAgentRegistry()` to get all 263 agent definitions
-3. Call `runHerald(context, registry)` to get the optimal roster
-4. Merge Herald's roster with this command's hardcoded lead agents (Herald adds, never removes leads)
-5. Deploy the merged roster per the command's normal parallel/sequential protocol
+Run: `npx thevoidforge herald --command /build --json`
+(Add `--focus "<topic>"` if the user provided `--focus`)
 
-**`--focus "topic"`** biases the Herald toward agents matching the topic. Examples: `--focus "security"`, `--focus "financial accuracy"`, `--focus "mobile UX"`.
+Parse the JSON output. The `roster` array contains agent IDs to deploy alongside this command's lead agents. If the command fails or returns an empty roster, use the hardcoded manifest below.
 
-**`--light`** skips the Herald entirely — uses only the command's hardcoded core roster.
-**`--solo`** skips both Herald and all sub-agents — lead agent only.
+**`--focus "topic"`** biases the Surfer toward agents matching the topic.
+**`--light`** skips the Surfer — uses only hardcoded core roster.
+**`--solo`** skips Surfer and all sub-agents — lead only.
 
 ## Context Setup
 1. Read `/logs/build-state.md` — if it exists, resume from current phase

package/.claude/commands/campaign.md

@@ -49,24 +49,22 @@ After planning mode completes, the user can run `/campaign` (no flags) to start
 
 ## Dynamic Dispatch (ADR-044)
 
-Opus scans `git diff --stat` and matches changed files against the `description` fields of all 263 agents in `.claude/agents/`. Matching specialists launch alongside the core agents below.
+Opus scans `git diff --stat` and matches changed files against the `description` fields of all 264 agents in `.claude/agents/`. Matching specialists launch alongside the core agents below.
 
 **Dispatch control:** `--light` skips dynamic dispatch (core only). `--solo` runs lead agent only.
 
-## Herald Pre-Scan (ADR-047)
+## Silver Surfer Pre-Scan (ADR-048)
 
-Before agent deployment, run the Herald to select the optimal roster:
+Before agent deployment, the Silver Surfer selects the optimal roster:
 
-1. Call `gatherHeraldContext('/campaign', '$ARGUMENTS', '<focus-if-provided>')` to collect codebase context
-2. Call `loadAgentRegistry()` to get all 263 agent definitions
-3. Call `runHerald(context, registry)` to get the optimal roster
-4. Merge Herald's roster with this command's hardcoded lead agents (Herald adds, never removes leads)
-5. Deploy the merged roster per the command's normal parallel/sequential protocol
+Run: `npx thevoidforge herald --command /campaign --json`
+(Add `--focus "<topic>"` if the user provided `--focus`)
 
-**`--focus "topic"`** biases the Herald toward agents matching the topic. Examples: `--focus "security"`, `--focus "financial accuracy"`, `--focus "mobile UX"`.
+Parse the JSON output. The `roster` array contains agent IDs to deploy alongside this command's lead agents. If the command fails or returns an empty roster, use the hardcoded manifest below.
 
-**`--light`** skips the Herald entirely — uses only the command's hardcoded core roster.
-**`--solo`** skips both Herald and all sub-agents — lead agent only.
+**`--focus "topic"`** biases the Surfer toward agents matching the topic.
+**`--light`** skips the Surfer — uses only hardcoded core roster.
+**`--solo`** skips Surfer and all sub-agents — lead only.
 
 ## Execution Mode (default)
 

package/.claude/commands/deploy.md

@@ -4,20 +4,18 @@
 
 Read `/docs/methods/DEVOPS_ENGINEER.md` for operating rules (see "Deploy Automation" section).
 
-## Herald Pre-Scan (ADR-047)
+## Silver Surfer Pre-Scan (ADR-048)
 
-Before agent deployment, run the Herald to select the optimal roster:
+Before agent deployment, the Silver Surfer selects the optimal roster:
 
-1. Call `gatherHeraldContext('/deploy', '$ARGUMENTS', '<focus-if-provided>')` to collect codebase context
-2. Call `loadAgentRegistry()` to get all 263 agent definitions
-3. Call `runHerald(context, registry)` to get the optimal roster
-4. Merge Herald's roster with this command's hardcoded lead agents (Herald adds, never removes leads)
-5. Deploy the merged roster per the command's normal parallel/sequential protocol
+Run: `npx thevoidforge herald --command /deploy --json`
+(Add `--focus "<topic>"` if the user provided `--focus`)
 
-**`--focus "topic"`** biases the Herald toward agents matching the topic. Examples: `--focus "security"`, `--focus "financial accuracy"`, `--focus "mobile UX"`.
+Parse the JSON output. The `roster` array contains agent IDs to deploy alongside this command's lead agents. If the command fails or returns an empty roster, use the hardcoded manifest below.
 
-**`--light`** skips the Herald entirely — uses only the command's hardcoded core roster.
-**`--solo`** skips both Herald and all sub-agents — lead agent only.
+**`--focus "topic"`** biases the Surfer toward agents matching the topic.
+**`--light`** skips the Surfer — uses only hardcoded core roster.
+**`--solo`** skips Surfer and all sub-agents — lead only.
 
 ## Context Setup
 1. Read PRD frontmatter for `deploy:` target (vps, vercel, railway, docker, static, cloudflare)

package/.claude/commands/devops.md

@@ -7,24 +7,22 @@
 
 ## Dynamic Dispatch (ADR-044)
 
-Opus scans `git diff --stat` and matches changed files against the `description` fields of all 263 agents in `.claude/agents/`. Matching specialists launch alongside the core agents below.
+Opus scans `git diff --stat` and matches changed files against the `description` fields of all 264 agents in `.claude/agents/`. Matching specialists launch alongside the core agents below.
 
 **Dispatch control:** `--light` skips dynamic dispatch (core only). `--solo` runs lead agent only.
 
-## Herald Pre-Scan (ADR-047)
+## Silver Surfer Pre-Scan (ADR-048)
 
-Before agent deployment, run the Herald to select the optimal roster:
+Before agent deployment, the Silver Surfer selects the optimal roster:
 
-1. Call `gatherHeraldContext('/devops', '$ARGUMENTS', '<focus-if-provided>')` to collect codebase context
-2. Call `loadAgentRegistry()` to get all 263 agent definitions
-3. Call `runHerald(context, registry)` to get the optimal roster
-4. Merge Herald's roster with this command's hardcoded lead agents (Herald adds, never removes leads)
-5. Deploy the merged roster per the command's normal parallel/sequential protocol
+Run: `npx thevoidforge herald --command /devops --json`
+(Add `--focus "<topic>"` if the user provided `--focus`)
 
-**`--focus "topic"`** biases the Herald toward agents matching the topic. Examples: `--focus "security"`, `--focus "financial accuracy"`, `--focus "mobile UX"`.
+Parse the JSON output. The `roster` array contains agent IDs to deploy alongside this command's lead agents. If the command fails or returns an empty roster, use the hardcoded manifest below.
 
-**`--light`** skips the Herald entirely — uses only the command's hardcoded core roster.
-**`--solo`** skips both Herald and all sub-agents — lead agent only.
+**`--focus "topic"`** biases the Surfer toward agents matching the topic.
+**`--light`** skips the Surfer — uses only hardcoded core roster.
+**`--solo`** skips Surfer and all sub-agents — lead only.
 
 ## Agent Deployment Manifest
 

package/.claude/commands/gauntlet.md

@@ -11,24 +11,22 @@ The Gauntlet tests everything. Every domain. Multiple rounds. Escalating intensi
 
 ## Dynamic Dispatch (ADR-044)
 
-Opus scans `git diff --stat` and matches changed files against the `description` fields of all 263 agents in `.claude/agents/`. Matching specialists launch alongside the core agents below.
+Opus scans `git diff --stat` and matches changed files against the `description` fields of all 264 agents in `.claude/agents/`. Matching specialists launch alongside the core agents below.
 
 **Dispatch control:** `--light` skips dynamic dispatch (core only). `--solo` runs lead agent only.
 
-## Herald Pre-Scan (ADR-047)
+## Silver Surfer Pre-Scan (ADR-048)
 
-Before agent deployment, run the Herald to select the optimal roster:
+Before agent deployment, the Silver Surfer selects the optimal roster:
 
-1. Call `gatherHeraldContext('/gauntlet', '$ARGUMENTS', '<focus-if-provided>')` to collect codebase context
-2. Call `loadAgentRegistry()` to get all 263 agent definitions
-3. Call `runHerald(context, registry)` to get the optimal roster
-4. Merge Herald's roster with this command's hardcoded lead agents (Herald adds, never removes leads)
-5. Deploy the merged roster per the command's normal parallel/sequential protocol
+Run: `npx thevoidforge herald --command /gauntlet --json`
+(Add `--focus "<topic>"` if the user provided `--focus`)
 
-**`--focus "topic"`** biases the Herald toward agents matching the topic. Examples: `--focus "security"`, `--focus "financial accuracy"`, `--focus "mobile UX"`.
+Parse the JSON output. The `roster` array contains agent IDs to deploy alongside this command's lead agents. If the command fails or returns an empty roster, use the hardcoded manifest below.
 
-**`--light`** skips the Herald entirely — uses only the command's hardcoded core roster.
-**`--solo`** skips both Herald and all sub-agents — lead agent only.
+**`--focus "topic"`** biases the Surfer toward agents matching the topic.
+**`--light`** skips the Surfer — uses only hardcoded core roster.
+**`--solo`** skips Surfer and all sub-agents — lead only.
 
 ## Round 1 — Discovery (parallel)
 

package/.claude/commands/qa.md

@@ -4,26 +4,24 @@
 
 ## Dynamic Dispatch (ADR-044)
 
-Opus scans `git diff --stat` and matches changed files against the `description` fields of all 263 agents in `.claude/agents/`. Matching specialists launch alongside the core agents below.
+Opus scans `git diff --stat` and matches changed files against the `description` fields of all 264 agents in `.claude/agents/`. Matching specialists launch alongside the core agents below.
 
 **Dispatch control:** `--light` skips dynamic dispatch (core only). `--solo` runs lead agent only.
 
 **Promoted agent:** **Constantine** `subagent_type: Constantine` runs on every `/qa` final pass — finds code that works by accident.
 
-## Herald Pre-Scan (ADR-047)
+## Silver Surfer Pre-Scan (ADR-048)
 
-Before agent deployment, run the Herald to select the optimal roster:
+Before agent deployment, the Silver Surfer selects the optimal roster:
 
-1. Call `gatherHeraldContext('/qa', '$ARGUMENTS', '<focus-if-provided>')` to collect codebase context
-2. Call `loadAgentRegistry()` to get all 263 agent definitions
-3. Call `runHerald(context, registry)` to get the optimal roster
-4. Merge Herald's roster with this command's hardcoded lead agents (Herald adds, never removes leads)
-5. Deploy the merged roster per the command's normal parallel/sequential protocol
+Run: `npx thevoidforge herald --command /qa --json`
+(Add `--focus "<topic>"` if the user provided `--focus`)
 
-**`--focus "topic"`** biases the Herald toward agents matching the topic. Examples: `--focus "security"`, `--focus "financial accuracy"`, `--focus "mobile UX"`.
+Parse the JSON output. The `roster` array contains agent IDs to deploy alongside this command's lead agents. If the command fails or returns an empty roster, use the hardcoded manifest below.
 
-**`--light`** skips the Herald entirely — uses only the command's hardcoded core roster.
-**`--solo`** skips both Herald and all sub-agents — lead agent only.
+**`--focus "topic"`** biases the Surfer toward agents matching the topic.
+**`--light`** skips the Surfer — uses only hardcoded core roster.
+**`--solo`** skips Surfer and all sub-agents — lead only.
 
 ## Context Setup
 1. Read `/logs/build-state.md` — understand current project state

package/.claude/commands/review.md

@@ -4,24 +4,22 @@
 
 ## Dynamic Dispatch (ADR-044)
 
-Opus scans `git diff --stat` and matches changed files against the `description` fields of all 263 agents in `.claude/agents/`. Matching specialists launch alongside the core agents below.
+Opus scans `git diff --stat` and matches changed files against the `description` fields of all 264 agents in `.claude/agents/`. Matching specialists launch alongside the core agents below.
 
 **Dispatch control:** `--light` skips dynamic dispatch (core only). `--solo` runs lead agent only.
 
-## Herald Pre-Scan (ADR-047)
+## Silver Surfer Pre-Scan (ADR-048)
 
-Before agent deployment, run the Herald to select the optimal roster:
+Before agent deployment, the Silver Surfer selects the optimal roster:
 
-1. Call `gatherHeraldContext('/review', '$ARGUMENTS', '<focus-if-provided>')` to collect codebase context
-2. Call `loadAgentRegistry()` to get all 263 agent definitions
-3. Call `runHerald(context, registry)` to get the optimal roster
-4. Merge Herald's roster with this command's hardcoded lead agents (Herald adds, never removes leads)
-5. Deploy the merged roster per the command's normal parallel/sequential protocol
+Run: `npx thevoidforge herald --command /review --json`
+(Add `--focus "<topic>"` if the user provided `--focus`)
 
-**`--focus "topic"`** biases the Herald toward agents matching the topic. Examples: `--focus "security"`, `--focus "financial accuracy"`, `--focus "mobile UX"`.
+Parse the JSON output. The `roster` array contains agent IDs to deploy alongside this command's lead agents. If the command fails or returns an empty roster, use the hardcoded manifest below.
 
-**`--light`** skips the Herald entirely — uses only the command's hardcoded core roster.
-**`--solo`** skips both Herald and all sub-agents — lead agent only.
+**`--focus "topic"`** biases the Surfer toward agents matching the topic.
+**`--light`** skips the Surfer — uses only hardcoded core roster.
+**`--solo`** skips Surfer and all sub-agents — lead only.
 
 ## Context Setup
 1. Read `/logs/build-state.md` — understand current project state

package/.claude/commands/security.md

@@ -4,24 +4,22 @@
 
 ## Dynamic Dispatch (ADR-044)
 
-Opus scans `git diff --stat` and matches changed files against the `description` fields of all 263 agents in `.claude/agents/`. Matching specialists launch alongside the core agents below.
+Opus scans `git diff --stat` and matches changed files against the `description` fields of all 264 agents in `.claude/agents/`. Matching specialists launch alongside the core agents below.
 
 **Dispatch control:** `--light` skips dynamic dispatch (core only). `--solo` runs lead agent only.
 
-## Herald Pre-Scan (ADR-047)
+## Silver Surfer Pre-Scan (ADR-048)
 
-Before agent deployment, run the Herald to select the optimal roster:
+Before agent deployment, the Silver Surfer selects the optimal roster:
 
-1. Call `gatherHeraldContext('/security', '$ARGUMENTS', '<focus-if-provided>')` to collect codebase context
-2. Call `loadAgentRegistry()` to get all 263 agent definitions
-3. Call `runHerald(context, registry)` to get the optimal roster
-4. Merge Herald's roster with this command's hardcoded lead agents (Herald adds, never removes leads)
-5. Deploy the merged roster per the command's normal parallel/sequential protocol
+Run: `npx thevoidforge herald --command /security --json`
+(Add `--focus "<topic>"` if the user provided `--focus`)
 
-**`--focus "topic"`** biases the Herald toward agents matching the topic. Examples: `--focus "security"`, `--focus "financial accuracy"`, `--focus "mobile UX"`.
+Parse the JSON output. The `roster` array contains agent IDs to deploy alongside this command's lead agents. If the command fails or returns an empty roster, use the hardcoded manifest below.
 
-**`--light`** skips the Herald entirely — uses only the command's hardcoded core roster.
-**`--solo`** skips both Herald and all sub-agents — lead agent only.
+**`--focus "topic"`** biases the Surfer toward agents matching the topic.
+**`--light`** skips the Surfer — uses only hardcoded core roster.
+**`--solo`** skips Surfer and all sub-agents — lead only.
 
 ## Context Setup
 1. Read `/logs/build-state.md` — understand current project state

package/.claude/commands/test.md

@@ -9,24 +9,22 @@
 
 ## Dynamic Dispatch (ADR-044)
 
-Opus scans `git diff --stat` and matches changed files against the `description` fields of all 263 agents in `.claude/agents/`. Matching specialists launch alongside the core agents below.
+Opus scans `git diff --stat` and matches changed files against the `description` fields of all 264 agents in `.claude/agents/`. Matching specialists launch alongside the core agents below.
 
 **Dispatch control:** `--light` skips dynamic dispatch (core only). `--solo` runs lead agent only.
 
-## Herald Pre-Scan (ADR-047)
+## Silver Surfer Pre-Scan (ADR-048)
 
-Before agent deployment, run the Herald to select the optimal roster:
+Before agent deployment, the Silver Surfer selects the optimal roster:
 
-1. Call `gatherHeraldContext('/test', '$ARGUMENTS', '<focus-if-provided>')` to collect codebase context
-2. Call `loadAgentRegistry()` to get all 263 agent definitions
-3. Call `runHerald(context, registry)` to get the optimal roster
-4. Merge Herald's roster with this command's hardcoded lead agents (Herald adds, never removes leads)
-5. Deploy the merged roster per the command's normal parallel/sequential protocol
+Run: `npx thevoidforge herald --command /test --json`
+(Add `--focus "<topic>"` if the user provided `--focus`)
 
-**`--focus "topic"`** biases the Herald toward agents matching the topic. Examples: `--focus "security"`, `--focus "financial accuracy"`, `--focus "mobile UX"`.
+Parse the JSON output. The `roster` array contains agent IDs to deploy alongside this command's lead agents. If the command fails or returns an empty roster, use the hardcoded manifest below.
 
-**`--light`** skips the Herald entirely — uses only the command's hardcoded core roster.
-**`--solo`** skips both Herald and all sub-agents — lead agent only.
+**`--focus "topic"`** biases the Surfer toward agents matching the topic.
+**`--light`** skips the Surfer — uses only hardcoded core roster.
+**`--solo`** skips Surfer and all sub-agents — lead only.
 
 ## Step 0 — Orient
 **Oracle** `subagent_type: Oracle` orients:

package/.claude/commands/ux.md

@@ -4,24 +4,22 @@
 
 ## Dynamic Dispatch (ADR-044)
 
-Opus scans `git diff --stat` and matches changed files against the `description` fields of all 263 agents in `.claude/agents/`. Matching specialists launch alongside the core agents below.
+Opus scans `git diff --stat` and matches changed files against the `description` fields of all 264 agents in `.claude/agents/`. Matching specialists launch alongside the core agents below.
 
 **Dispatch control:** `--light` skips dynamic dispatch (core only). `--solo` runs lead agent only.
 
-## Herald Pre-Scan (ADR-047)
+## Silver Surfer Pre-Scan (ADR-048)
 
-Before agent deployment, run the Herald to select the optimal roster:
+Before agent deployment, the Silver Surfer selects the optimal roster:
 
-1. Call `gatherHeraldContext('/ux', '$ARGUMENTS', '<focus-if-provided>')` to collect codebase context
-2. Call `loadAgentRegistry()` to get all 263 agent definitions
-3. Call `runHerald(context, registry)` to get the optimal roster
-4. Merge Herald's roster with this command's hardcoded lead agents (Herald adds, never removes leads)
-5. Deploy the merged roster per the command's normal parallel/sequential protocol
+Run: `npx thevoidforge herald --command /ux --json`
+(Add `--focus "<topic>"` if the user provided `--focus`)
 
-**`--focus "topic"`** biases the Herald toward agents matching the topic. Examples: `--focus "security"`, `--focus "financial accuracy"`, `--focus "mobile UX"`.
+Parse the JSON output. The `roster` array contains agent IDs to deploy alongside this command's lead agents. If the command fails or returns an empty roster, use the hardcoded manifest below.
 
-**`--light`** skips the Herald entirely — uses only the command's hardcoded core roster.
-**`--solo`** skips both Herald and all sub-agents — lead agent only.
+**`--focus "topic"`** biases the Surfer toward agents matching the topic.
+**`--light`** skips the Surfer — uses only hardcoded core roster.
+**`--solo`** skips Surfer and all sub-agents — lead only.
 
 ## Context Setup
 1. Read `/logs/build-state.md` — understand current project state

package/CHANGELOG.md

@@ -6,6 +6,30 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/), and this
 
 ---
 
+## [23.6.1] - 2026-04-12
+
+### Fixed
+- **30+ stale "263 agents" references** — updated to 264 across 13 command files, 8 method docs, HOLOCRON.md, AGENT_CLASSIFICATION.md, ROADMAP.md, wizard UI, herald.ts, SUB_AGENTS.md, and NAMING_REGISTRY.md. Gauntlet caught incomplete count propagation from Silver Surfer addition.
+- **herald.ts ADR reference** — ADR-047 → ADR-048
+
+---
+
+## [23.6.0] - 2026-04-12
+
+### The Silver Surfer (ADR-048, Campaign 38)
+
+### Added
+- **Silver Surfer** (Norrin Radd) — agent #264. Herald of Galactus. Pre-scan dispatch that reads codebase context and selects the optimal agent roster via Haiku.
+- **`voidforge herald` CLI subcommand** — the invocation bridge. Runs the Silver Surfer pre-scan from the command line: `npx thevoidforge herald --command /review --json`. Outputs JSON roster for Claude to deploy.
+- Silver Surfer agent definition with Haiku tier, scout tools, and dispatch tags.
+
+### Changed
+- All 14 major commands updated: "Herald Pre-Scan (ADR-047)" replaced with "Silver Surfer Pre-Scan (ADR-048)" using actual CLI invocation instead of pseudocode.
+- Agent count: 263 → 264.
+- CLAUDE.md and NAMING_REGISTRY.md updated with Silver Surfer identity.
+
+---
+
 ## [23.5.4] - 2026-04-12
 
 ### Fixed

package/CLAUDE.md

@@ -234,7 +234,7 @@ See `/docs/methods/MUSTER.md` for the full Muster Protocol.
 | Deploy Wizard | **Haku** (Anime) | Browser-based deploy wizard, infrastructure provisioning |
 | Setup Wizard | **Gandalf** (Tolkien) | Project scaffolding, initialization, dependency setup |
 
-263 sub-agent names in `/docs/NAMING_REGISTRY.md`. No duplicates across active sessions. All agents materialized as subagent definitions in `.claude/agents/` (ADR-044).
+264 sub-agent names in `/docs/NAMING_REGISTRY.md`. No duplicates across active sessions. All agents materialized as subagent definitions in `.claude/agents/` (ADR-044). Silver Surfer (ADR-048) dispatches the optimal roster via Haiku pre-scan.
 
 ## Distribution
 

package/HOLOCRON.md

@@ -24,7 +24,7 @@
 
 ### What VoidForge Is
 
-VoidForge is a **methodology framework** for building full-stack applications with Claude Code. It's not a code template — it's a *process* template. Drop in a Product Requirements Document, and a named team of 263 AI agents across 9 fictional universes builds your application through a 13-phase protocol.
+VoidForge is a **methodology framework** for building full-stack applications with Claude Code. It's not a code template — it's a *process* template. Drop in a Product Requirements Document, and a named team of 264 AI agents across 9 fictional universes builds your application through a 13-phase protocol.
 
 **From nothing, everything.**
 
@@ -82,7 +82,7 @@ Every tier includes:
 - **28 slash commands** — `/prd`, `/blueprint`, `/build`, `/qa`, `/test`, `/security`, `/ux`, `/review`, `/deploy`, `/devops`, `/architect`, `/assess`, `/git`, `/void`, `/vault`, `/thumper`, `/assemble`, `/gauntlet`, `/campaign`, `/imagine`, `/debrief`, `/dangerroom`, `/cultivation`, `/grow`, `/current`, `/treasury`, `/portfolio`, `/ai`
 - **13-phase build protocol** — PRD to production with verification gates
 - **18 specialist agent protocols** — Each lead has behavioral directives and a sub-agent roster
-- **263 named characters** — From Tolkien, Marvel, DC, Star Wars, Star Trek, Dune, Anime, Cosmere, and Foundation — each materialized as a subagent definition in `.claude/agents/`
+- **264 named characters** — From Tolkien, Marvel, DC, Star Wars, Star Trek, Dune, Anime, Cosmere, and Foundation — each materialized as a subagent definition in `.claude/agents/`
 - **35 code patterns** — Reference implementations with framework adaptations (including E2E testing)
 - **No Stubs Doctrine** — Zero placeholder code. Every file does what it claims. Enforced across all method docs.
 - **E2E browser testing** — Playwright + axe-core. Agents take screenshots, capture console errors, and interact with running applications.
@@ -366,7 +366,7 @@ Each lead has a deep bench. Here are some standouts:
 - **Idaho** keeps the connection alive — the eternal ghola who always returns
 - **Thufir** parses every signal — Mentat precision, a million calculations per second
 
-The full roster of 263 characters lives in `docs/NAMING_REGISTRY.md`. Each agent is materialized as a subagent definition in `.claude/agents/` with model tiering (Opus leads, Sonnet specialists, Haiku scouts) and tool restrictions (Builder, Reviewer, Scout, Adversarial). See `docs/AGENT_CLASSIFICATION.md` for the full manifest.
+The full roster of 264 characters lives in `docs/NAMING_REGISTRY.md`. Each agent is materialized as a subagent definition in `.claude/agents/` with model tiering (Opus leads, Sonnet specialists, Haiku scouts) and tool restrictions (Builder, Reviewer, Scout, Adversarial). See `docs/AGENT_CLASSIFICATION.md` for the full manifest.
 
 ### How Handoffs Work
 

package/VERSION.md

@@ -1,6 +1,6 @@
 # Version
 
-**Current:** 23.5.4
+**Current:** 23.6.1
 
 ## Versioning Scheme
 
@@ -14,6 +14,8 @@ This project uses [Semantic Versioning](https://semver.org/):
 
 | Version | Date | Summary |
 |---------|------|---------|
+| 23.6.1 | 2026-04-12 | Gauntlet fix: 30+ stale "263 agents" → 264 across commands, methods, docs |
+| 23.6.0 | 2026-04-12 | The Silver Surfer — Herald invocation bridge, CLI subcommand, agent #264, end-to-end dispatch pipeline. ADR-048. Campaign 38. |
 | 23.5.4 | 2026-04-12 | Command-doc sync: build.md Phase 12.75, ux.md screenshots, qa.md dynamic counts |
 | 23.5.3 | 2026-04-12 | Fix 201 broken subagent_type refs (filename→YAML name) + /campaign as default start command |
 | 23.5.2 | 2026-04-12 | /void auto-cleanup ~/.claude/ duplicates + git init stack trace fix |

package/docs/AGENT_CLASSIFICATION.md

@@ -7,13 +7,13 @@
 
 | Metric | Count |
 |--------|-------|
-| **Total agents** | **263** |
+| **Total agents** | **264** |
 | Leads (Opus/inherit, Builder) | 20 |
 | Specialists (Sonnet, Reviewer) | 190 |
 | Scouts (Haiku, Scout/Reviewer) | 38 |
 | Adversarial (Sonnet, Adversarial) | 15 |
 
-**Note:** The authoritative count from NAMING_REGISTRY.md is 263. All methodology docs updated in M7. ADR-044 and ADR-042 retain their original "259" count as historical records.
+**Note:** The authoritative count from NAMING_REGISTRY.md is 264. All methodology docs updated in M7. ADR-044 and ADR-042 retain their original "259" count as historical records.
 
 ## Tiers (Model Assignment)
 
@@ -468,7 +468,7 @@
 | Dune | 1 | 14 | 4 | 2 | 21 |
 | Cosmere | 2 | 16 | 0 | 0 | 18 |
 | Foundation | 1 | 10 | 1 | 1 | 13 |
-| **TOTAL** | **20** | **190** | **38** | **15** | **263** |
+| **TOTAL** | **20** | **190** | **39** | **15** | **264** |
 
 ## Riker's Review — Edge Case Challenges
 

package/docs/NAMING_REGISTRY.md

@@ -113,8 +113,9 @@
 28. Jarvis — AI assistant, status reporting, progress summaries
 
 29. Thanos — **Lead: The Gauntlet.** Comprehensive multi-round review. Tests every domain, every stone. "I am inevitable."
+30. Silver Surfer (Norrin Radd) — **The Herald.** Pre-scan dispatch — reads codebase and selects optimal agent roster via Haiku. Scouts ahead, summons the right team. "All that you know is at an end."
 
-**Reserved:** Stark (lead — backend), Coulson (lead — release), Fury (lead — the initiative), Thanos (lead — the gauntlet)
+**Reserved:** Stark (lead — backend), Coulson (lead — release), Fury (lead — the initiative), Thanos (lead — the gauntlet), Silver Surfer (herald — dispatch)
 
 ---
 

package/docs/methods/CAMPAIGN.md

@@ -239,7 +239,7 @@ User confirms, redirects, or overrides. On confirm → Step 4.
 
 **Post-infrastructure enforcement gate:** For infrastructure campaigns (deploy targets, CI/CD, monitoring, staging environments): after the infrastructure is provisioned, run `/architect --plan` to verify workflow enforcement gates exist — not just infrastructure existence. Infrastructure without process gates is incomplete.
 
-**Dispatch model (ADR-044):** Per-mission `/assemble` runs SHOULD dispatch phases to sub-agents per `SUB_AGENTS.md` "Parallel Agent Standard." Agents are launched as named subagent types defined in `.claude/agents/` with description-driven dispatch — Opus scans `git diff --stat` and matches changed files against agent descriptions to auto-select specialists. The campaign orchestrator (main thread) manages the mission sequence, inter-mission gates, and campaign state — it does NOT perform inline code analysis. Pass findings summaries between missions, not raw code. See `docs/AGENT_CLASSIFICATION.md` for the full 263-agent manifest. (Field report #270)
+**Dispatch model (ADR-044):** Per-mission `/assemble` runs SHOULD dispatch phases to sub-agents per `SUB_AGENTS.md` "Parallel Agent Standard." Agents are launched as named subagent types defined in `.claude/agents/` with description-driven dispatch — Opus scans `git diff --stat` and matches changed files against agent descriptions to auto-select specialists. The campaign orchestrator (main thread) manages the mission sequence, inter-mission gates, and campaign state — it does NOT perform inline code analysis. Pass findings summaries between missions, not raw code. See `docs/AGENT_CLASSIFICATION.md` for the full 264-agent manifest. (Field report #270)
 
 ### Campaign-Mode Pipeline
 

package/docs/methods/GAUNTLET.md

@@ -18,7 +18,7 @@
 - When the project has significant attack surface (auth, payments, user data, WebSocket, file uploads)
 - Before a public launch or investor demo
 
-**Dispatch model:** All Gauntlet rounds MUST dispatch to sub-agents per `SUB_AGENTS.md` "Parallel Agent Standard." Agents are launched as named subagent types defined in `.claude/agents/` with model tiering (Opus leads, Sonnet specialists, Haiku scouts) and tool restrictions. The main thread manages rounds, triages findings between rounds, and applies fixes — it does NOT read source files or analyze code inline. Each round launches agents in waves of 3 (max concurrent). Findings pass between rounds as summary tables, not raw code. See `docs/AGENT_CLASSIFICATION.md` for the full 263-agent manifest. (Field report #270)
+**Dispatch model:** All Gauntlet rounds MUST dispatch to sub-agents per `SUB_AGENTS.md` "Parallel Agent Standard." Agents are launched as named subagent types defined in `.claude/agents/` with model tiering (Opus leads, Sonnet specialists, Haiku scouts) and tool restrictions. The main thread manages rounds, triages findings between rounds, and applies fixes — it does NOT read source files or analyze code inline. Each round launches agents in waves of 3 (max concurrent). Findings pass between rounds as summary tables, not raw code. See `docs/AGENT_CLASSIFICATION.md` for the full 264-agent manifest. (Field report #270)
 
 **When NOT to use /gauntlet:**
 - During active development (use `/assemble` instead — it includes building)

package/docs/methods/MUSTER.md

@@ -143,7 +143,7 @@ Synthesize all findings:
 
 ## Subagent Definitions (ADR-044)
 
-Muster agents are now launched as named subagent types defined in `.claude/agents/`. Instead of inline prompts, each agent invocation uses `subagent_type: {agent-id}` to reference a materialized definition with model tiering (Opus leads, Sonnet specialists, Haiku scouts) and tool restrictions. See `docs/AGENT_CLASSIFICATION.md` for the full 263-agent classification manifest.
+Muster agents are now launched as named subagent types defined in `.claude/agents/`. Instead of inline prompts, each agent invocation uses `subagent_type: {agent-id}` to reference a materialized definition with model tiering (Opus leads, Sonnet specialists, Haiku scouts) and tool restrictions. See `docs/AGENT_CLASSIFICATION.md` for the full 264-agent classification manifest.
 
 ## Cost Awareness
 

package/docs/methods/QA_ENGINEER.md

@@ -27,7 +27,7 @@
 
 ## Dynamic Dispatch (ADR-044)
 
-Agent dispatch is now description-driven. When Opus processes a command, it scans `git diff --stat` and matches changed files against the `description` fields of all 263 agents in `.claude/agents/`. Matching specialists launch automatically alongside core agents. No static dispatch tables needed.
+Agent dispatch is now description-driven. When Opus processes a command, it scans `git diff --stat` and matches changed files against the `description` fields of all 264 agents in `.claude/agents/`. Matching specialists launch automatically alongside core agents. No static dispatch tables needed.
 
 See `docs/AGENT_CLASSIFICATION.md` for the full classification and `docs/adrs/ADR-044-subagent-materialization.md` for the architecture.
 

package/docs/methods/SECURITY_AUDITOR.md

@@ -28,7 +28,7 @@
 
 ## Dynamic Dispatch (ADR-044)
 
-Agent dispatch is now description-driven. When Opus processes a command, it scans `git diff --stat` and matches changed files against the `description` fields of all 263 agents in `.claude/agents/`. Matching specialists launch automatically alongside core agents. No static dispatch tables needed.
+Agent dispatch is now description-driven. When Opus processes a command, it scans `git diff --stat` and matches changed files against the `description` fields of all 264 agents in `.claude/agents/`. Matching specialists launch automatically alongside core agents. No static dispatch tables needed.
 
 See `docs/AGENT_CLASSIFICATION.md` for the full classification and `docs/adrs/ADR-044-subagent-materialization.md` for the architecture.
 

package/docs/methods/SUB_AGENTS.md

@@ -10,7 +10,7 @@ Parallelize development across multiple Claude Code sessions. Each session runs
 
 **All orchestration uses the build journal.** Every delegation, handoff, and resolution is logged to `/logs/handoffs.md`. Agents read journal files to recover context. See `/docs/methods/BUILD_JOURNAL.md`.
 
-**Full character roster: `/docs/NAMING_REGISTRY.md`** — 263 named characters across 9 universes. No duplicates allowed. All agents materialized as subagent definitions in `.claude/agents/`.
+**Full character roster: `/docs/NAMING_REGISTRY.md`** — 264 named characters across 9 universes. No duplicates allowed. All agents materialized as subagent definitions in `.claude/agents/`.
 
 ---
 
@@ -176,7 +176,7 @@ Users can create project-specific sub-agents that carry domain knowledge. Define
 
 ## Subagent Definitions (ADR-044)
 
-All 263 agents are materialized as `.claude/agents/{name}.md` files. Commands now use `subagent_type: {agent-id}` instead of inline prompts. Each definition contains the agent's identity, behavioral directives, domain expertise, and output format.
+All 264 agents are materialized as `.claude/agents/{name}.md` files. Commands now use `subagent_type: {agent-id}` instead of inline prompts. Each definition contains the agent's identity, behavioral directives, domain expertise, and output format.
 
 ### The `.claude/agents/` Directory
 
@@ -206,7 +206,7 @@ Leads inherit the main session's model (Opus). Specialists run on Sonnet for cos
 Commands no longer use static dispatch tables (the old ADR-042 Cross-Domain Triggers). Instead, when Opus processes a command, it:
 
 1. Scans `git diff --stat` to identify changed files
-2. Matches changed file paths against the `description` fields of all 263 agents
+2. Matches changed file paths against the `description` fields of all 264 agents
 3. Launches matching specialists automatically alongside core agents
 
 This means a security audit that touches database migrations automatically picks up Spock (schema) without a hardcoded trigger table. The descriptions in `.claude/agents/*.md` ARE the dispatch rules.

package/docs/methods/SYSTEMS_ARCHITECT.md

@@ -25,7 +25,7 @@
 
 ## Dynamic Dispatch (ADR-044)
 
-Agent dispatch is now description-driven. When Opus processes a command, it scans `git diff --stat` and matches changed files against the `description` fields of all 263 agents in `.claude/agents/`. Matching specialists launch automatically alongside core agents. No static dispatch tables needed.
+Agent dispatch is now description-driven. When Opus processes a command, it scans `git diff --stat` and matches changed files against the `description` fields of all 264 agents in `.claude/agents/`. Matching specialists launch automatically alongside core agents. No static dispatch tables needed.
 
 See `docs/AGENT_CLASSIFICATION.md` for the full classification and `docs/adrs/ADR-044-subagent-materialization.md` for the architecture.
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "thevoidforge-methodology",
-  "version": "23.5.4",
+  "version": "23.6.1",
   "description": "VoidForge methodology — agents, commands, methods, patterns.",
   "license": "MIT",
   "files": [