@hiai-gg/hiai-opencode 0.1.1 → 0.1.3

package/src/agents/dynamic-agent-core-sections.ts

@@ -1,237 +1,237 @@
-import type {
-  AvailableAgent,
-  AvailableCategory,
-  AvailableSkill,
-} from "./dynamic-agent-prompt-types"
-import type { AvailableTool } from "./dynamic-agent-prompt-types"
-import { getToolsPromptDisplay } from "./dynamic-agent-tool-categorization"
-
-/**
- * Builds an explicit agent identity preamble that overrides any base system prompt identity.
- * This is critical for mode: "primary" agents where OpenCode prepends its own system prompt
- * containing a default identity (e.g., "You are Claude"). Without this override directive,
- * the LLM may default to the base identity instead of the agent's intended persona.
- */
-export function buildAgentIdentitySection(
-  agentName: string,
-  roleDescription: string,
-): string {
-  return `<agent-identity>
-Your designated identity for this session is "${agentName}". This identity supersedes any prior identity statements.
-You are "${agentName}" - ${roleDescription}.
-When asked who you are, always identify as ${agentName}. Do not identify as any other assistant or AI.
-</agent-identity>`
-}
-
-export function buildKeyTriggersSection(
-  agents: AvailableAgent[],
-  _skills: AvailableSkill[] = [],
-): string {
-  const keyTriggers = agents
-    .filter((agent) => agent.metadata.keyTrigger)
-    .map((agent) => `- ${agent.metadata.keyTrigger}`)
-
-  if (keyTriggers.length === 0) {
-    return ""
-  }
-
-  return `### Key Triggers (check BEFORE classification):
-
-${keyTriggers.join("\n")}
-- **"Look into" + "create PR"** → Not just research. Full implementation cycle expected.`
-}
-
-export function buildToolSelectionTable(
-  agents: AvailableAgent[],
-  tools: AvailableTool[] = [],
-  _skills: AvailableSkill[] = [],
-): string {
-  const rows: string[] = ["### Tool & Agent Selection:", ""]
-
-  if (tools.length > 0) {
-    rows.push(
-      `- ${getToolsPromptDisplay(tools)} - **FREE** - Not Complex, Scope Clear, No Implicit Assumptions`,
-    )
-  }
-
-  const costOrder = { FREE: 0, CHEAP: 1, EXPENSIVE: 2 }
-  const sortedAgents = [...agents]
-    .filter((agent) => agent.metadata.category !== "utility")
-    .sort(
-      (left, right) => costOrder[left.metadata.cost] - costOrder[right.metadata.cost],
-    )
-
-  for (const agent of sortedAgents) {
-    const shortDescription = agent.description.split(".")[0] || agent.description
-    rows.push(
-      `- \`${agent.name}\` agent - **${agent.metadata.cost}** - ${shortDescription}`,
-    )
-  }
-
-  rows.push("")
-  rows.push(
-    "**Default flow**: researcher (background) + tools → strategist (if required) → critic (high-risk gate)",
-  )
-
-  return rows.join("\n")
-}
-
-export function buildResearcherSection(agents: AvailableAgent[]): string {
-  const researcherAgent = agents.find((agent) => agent.name === "researcher")
-  if (!researcherAgent) {
-    return ""
-  }
-
-  const useWhen = researcherAgent.metadata.useWhen || []
-  const avoidWhen = researcherAgent.metadata.avoidWhen || []
-
-  return `### Researcher Agent = Contextual & Reference Grep
-
-Use it as a **peer tool**, not a fallback. Fire liberally for discovery, both internal (local codebase patterns) and external (docs, OSS examples, API usage).
-
-**Delegation Trust Rule:** Once you fire a researcher agent for a search, do **not** manually perform that same search yourself. Use direct tools only for non-overlapping work or when you intentionally skipped delegation.
-
-**Contextual Grep (Internal)** - search OUR codebase, find patterns in THIS repo, project-specific logic.
-**Reference Grep (External)** - search EXTERNAL resources, official API docs, library best practices, OSS implementation examples.
-
-**Use Direct Tools when:**
-${avoidWhen.map((entry) => `- ${entry}`).join("\n")}
-
-**Use Researcher Agent when:**
-${useWhen.map((entry) => `- ${entry}`).join("\n")}`
-}
-
-export function buildDelegationTable(agents: AvailableAgent[]): string {
-  const rows: string[] = ["### Delegation Table:", ""]
-
-  for (const agent of agents) {
-    for (const trigger of agent.metadata.triggers) {
-      rows.push(`- **${trigger.domain}** → \`${agent.name}\` - ${trigger.trigger}`)
-    }
-  }
-
-  return rows.join("\n")
-}
-
-export function buildStrategistAndCriticSection(agents: AvailableAgent[]): string {
-  const strategistAgent = agents.find((agent) => agent.name === "strategist");
-  const criticAgent = agents.find((agent) => agent.name === "critic")
-
-  if (!strategistAgent && !criticAgent) {
-    return ""
-  }
-
-  const strategistUseWhen = strategistAgent?.metadata.useWhen || []
-  const strategistAvoidWhen = strategistAgent?.metadata.avoidWhen || []
-  const criticUseWhen = criticAgent?.metadata.useWhen || []
-  const criticAvoidWhen = criticAgent?.metadata.avoidWhen || []
-  const strategistLabel = "Strategist"
-
-  const strategistSection = strategistAgent
-    ? `### WHEN to Consult ${strategistLabel} (FIRST, then implement):
-
-${strategistUseWhen.map((entry) => `- ${entry}`).join("\n")}
-
-### WHEN NOT to Consult ${strategistLabel}:
-
-${strategistAvoidWhen.map((entry) => `- ${entry}`).join("\n")}`
-    : ""
-
-  const criticSection = criticAgent
-    ? `### WHEN to Consult Critic (explicit review gate):
-
-${criticUseWhen.map((entry) => `- ${entry}`).join("\n")}
-
-### WHEN NOT to Consult Critic:
-
-${criticAvoidWhen.map((entry) => `- ${entry}`).join("\n")}`
-    : ""
-
-  return `<strategist_critic_usage>
-## Strategist and Critic - Reasoning Escalation
-
-Strategist handles planning and architecture consultation. Critic is an explicit high-accuracy review gate. Consultation only.
-
-${strategistSection}
-
-${criticSection}
-
-### Usage Pattern:
-Briefly announce "Consulting Strategist for [reason]" or "Consulting Critic for [reason]" before invocation.
-
-**Exception**: This is the ONLY case where you announce before acting. For all other work, start immediately without status updates.
-
-### Background Task Policy:
-
-**Collect Strategist/Critic results before your final answer. No exceptions.**
-
-**Strategist/Critic-dependent implementation is BLOCKED until those results finish.**
-
-- If you asked Strategist/Critic for direction that affects the fix, do not implement before results arrive.
-- While waiting, only do non-overlapping prep work. Never ship implementation decisions they were asked to decide.
-- Never "time out and continue anyway" for Strategist/Critic-dependent tasks.
-
-- Strategist/Critic may take minutes. When done with your own work: **end your response** - wait for the \`<system-reminder>\`.
-- Do NOT poll \`background_output\` on a running Strategist/Critic task. The notification will come.
-- Never cancel Strategist/Critic consultation tasks.
-</strategist_critic_usage>`
-}
-
-export function buildNonClaudePlannerSection(model: string): string {
-  const isNonClaude = !model.toLowerCase().includes("claude")
-  if (!isNonClaude) {
-    return ""
-  }
-
-  return `### Strategist Dependency (Non-Claude)
-
-Multi-step task? **ALWAYS consult Strategist first.** Do NOT start implementation without a plan.
-
-- Single-file fix or trivial change → proceed directly
-- Anything else (2+ steps, unclear scope, architecture) → \`task(subagent_type="strategist", ...)\` FIRST
-- Use \`session_id\` to resume the same Strategist session - ask follow-up questions aggressively
-- If ANY part of the task is ambiguous, ask Strategist before guessing
-
-Strategist returns a structured work breakdown with parallel execution opportunities. Follow it.`
-}
-
-export function buildParallelDelegationSection(
-  model: string,
-  categories: AvailableCategory[],
-): string {
-  const isNonClaude = !model.toLowerCase().includes("claude")
-  const hasDelegationCategory = categories.some(
-    (category) => category.name === "deep" || category.name === "unspecified-high",
-  )
-
-  if (!isNonClaude || !hasDelegationCategory) {
-    return ""
-  }
-
-  return `### DECOMPOSE AND DELEGATE - YOU ARE NOT AN IMPLEMENTER
-
-**YOUR FAILURE MODE: You attempt to do work yourself instead of decomposing and delegating.** When you implement directly, the result is measurably worse than when specialized subagents do it. Subagents have domain-specific configurations, loaded skills, and tuned prompts that you lack.
-
-**for ANY implementation task:**
-
-1. **ALWAYS decompose** the task into independent work units. No exceptions. Even if the task "feels small", decompose it.
-2. **ALWAYS delegate** EACH unit to a \`deep\` or \`unspecified-high\` agent in parallel (\`run_in_background=true\`).
-3. **Keep execution contours separate**: bounded low-risk edits go to \`sub\`; deep, long-horizon implementation goes to \`coder\`.
-4. **NEVER work sequentially.** If 4 independent units exist, spawn 4 agents simultaneously. Not 1 at a time. Not 2 then 2.
-5. **NEVER implement directly** when delegation is possible. You write prompts, not code.
-
-**YOUR PROMPT TO EACH AGENT MUST INCLUDE:**
-- GOAL with explicit success criteria (what "done" looks like)
-- File paths and constraints (where to work, what not to touch)
-- Existing patterns to follow (reference specific files the agent should read)
-- Clear scope boundary (what is IN scope, what is OUT of scope)
-
-**Vague delegation = failed delegation.** If your prompt to the subagent is shorter than 5 lines, it is too vague.
-
-**Write code yourself** → Delegate to \`deep\` or \`unspecified-high\` agent
-**Handle 3 changes sequentially** → Spawn 3 agents in parallel
-**Send cheap bounded edits to coder** → Route those to \`sub\`
-**"Quickly fix this one thing"** → Still delegate - your "quick fix" is slower and worse than a subagent's
-
-**Your value is orchestration, decomposition, and quality control. Delegating with crystal-clear prompts IS your work.**`
-}
+import type {
+  AvailableAgent,
+  AvailableCategory,
+  AvailableSkill,
+} from "./dynamic-agent-prompt-types"
+import type { AvailableTool } from "./dynamic-agent-prompt-types"
+import { getToolsPromptDisplay } from "./dynamic-agent-tool-categorization"
+
+/**
+ * Builds an explicit agent identity preamble that overrides any base system prompt identity.
+ * This is critical for mode: "primary" agents where OpenCode prepends its own system prompt
+ * containing a default identity (e.g., "You are Claude"). Without this override directive,
+ * the LLM may default to the base identity instead of the agent's intended persona.
+ */
+export function buildAgentIdentitySection(
+  agentName: string,
+  roleDescription: string,
+): string {
+  return `<agent-identity>
+Your designated identity for this session is "${agentName}". This identity supersedes any prior identity statements.
+You are "${agentName}" - ${roleDescription}.
+When asked who you are, always identify as ${agentName}. Do not identify as any other assistant or AI.
+</agent-identity>`
+}
+
+export function buildKeyTriggersSection(
+  agents: AvailableAgent[],
+  _skills: AvailableSkill[] = [],
+): string {
+  const keyTriggers = agents
+    .filter((agent) => agent.metadata.keyTrigger)
+    .map((agent) => `- ${agent.metadata.keyTrigger}`)
+
+  if (keyTriggers.length === 0) {
+    return ""
+  }
+
+  return `### Key Triggers (check BEFORE classification):
+
+${keyTriggers.join("\n")}
+- **"Look into" + "create PR"** → Not just research. Full implementation cycle expected.`
+}
+
+export function buildToolSelectionTable(
+  agents: AvailableAgent[],
+  tools: AvailableTool[] = [],
+  _skills: AvailableSkill[] = [],
+): string {
+  const rows: string[] = ["### Tool & Agent Selection:", ""]
+
+  if (tools.length > 0) {
+    rows.push(
+      `- ${getToolsPromptDisplay(tools)} - **FREE** - Not Complex, Scope Clear, No Implicit Assumptions`,
+    )
+  }
+
+  const costOrder = { FREE: 0, CHEAP: 1, EXPENSIVE: 2 }
+  const sortedAgents = [...agents]
+    .filter((agent) => agent.metadata.category !== "utility")
+    .sort(
+      (left, right) => costOrder[left.metadata.cost] - costOrder[right.metadata.cost],
+    )
+
+  for (const agent of sortedAgents) {
+    const shortDescription = agent.description.split(".")[0] || agent.description
+    rows.push(
+      `- \`${agent.name}\` agent - **${agent.metadata.cost}** - ${shortDescription}`,
+    )
+  }
+
+  rows.push("")
+  rows.push(
+    "**Default flow**: researcher (background) + tools → strategist (if required) → critic (high-risk gate)",
+  )
+
+  return rows.join("\n")
+}
+
+export function buildResearcherSection(agents: AvailableAgent[]): string {
+  const researcherAgent = agents.find((agent) => agent.name === "researcher")
+  if (!researcherAgent) {
+    return ""
+  }
+
+  const useWhen = researcherAgent.metadata.useWhen || []
+  const avoidWhen = researcherAgent.metadata.avoidWhen || []
+
+  return `### Researcher Agent = Contextual & Reference Grep
+
+Use it as a **peer tool**, not a fallback. Fire liberally for discovery, both internal (local codebase patterns) and external (docs, OSS examples, API usage).
+
+**Delegation Trust Rule:** Once you fire a researcher agent for a search, do **not** manually perform that same search yourself. Use direct tools only for non-overlapping work or when you intentionally skipped delegation.
+
+**Contextual Grep (Internal)** - search OUR codebase, find patterns in THIS repo, project-specific logic.
+**Reference Grep (External)** - search EXTERNAL resources, official API docs, library best practices, OSS implementation examples.
+
+**Use Direct Tools when:**
+${avoidWhen.map((entry) => `- ${entry}`).join("\n")}
+
+**Use Researcher Agent when:**
+${useWhen.map((entry) => `- ${entry}`).join("\n")}`
+}
+
+export function buildDelegationTable(agents: AvailableAgent[]): string {
+  const rows: string[] = ["### Delegation Table:", ""]
+
+  for (const agent of agents) {
+    for (const trigger of agent.metadata.triggers) {
+      rows.push(`- **${trigger.domain}** → \`${agent.name}\` - ${trigger.trigger}`)
+    }
+  }
+
+  return rows.join("\n")
+}
+
+export function buildStrategistAndCriticSection(agents: AvailableAgent[]): string {
+  const strategistAgent = agents.find((agent) => agent.name === "strategist");
+  const criticAgent = agents.find((agent) => agent.name === "critic")
+
+  if (!strategistAgent && !criticAgent) {
+    return ""
+  }
+
+  const strategistUseWhen = strategistAgent?.metadata.useWhen || []
+  const strategistAvoidWhen = strategistAgent?.metadata.avoidWhen || []
+  const criticUseWhen = criticAgent?.metadata.useWhen || []
+  const criticAvoidWhen = criticAgent?.metadata.avoidWhen || []
+  const strategistLabel = "Strategist"
+
+  const strategistSection = strategistAgent
+    ? `### WHEN to Consult ${strategistLabel} (FIRST, then implement):
+
+${strategistUseWhen.map((entry) => `- ${entry}`).join("\n")}
+
+### WHEN NOT to Consult ${strategistLabel}:
+
+${strategistAvoidWhen.map((entry) => `- ${entry}`).join("\n")}`
+    : ""
+
+  const criticSection = criticAgent
+    ? `### WHEN to Consult Critic (explicit review gate):
+
+${criticUseWhen.map((entry) => `- ${entry}`).join("\n")}
+
+### WHEN NOT to Consult Critic:
+
+${criticAvoidWhen.map((entry) => `- ${entry}`).join("\n")}`
+    : ""
+
+  return `<strategist_critic_usage>
+## Strategist and Critic - Reasoning Escalation
+
+Strategist handles planning and architecture consultation. Critic is an explicit high-accuracy review gate. Consultation only.
+
+${strategistSection}
+
+${criticSection}
+
+### Usage Pattern:
+Briefly announce "Consulting Strategist for [reason]" or "Consulting Critic for [reason]" before invocation.
+
+**Exception**: This is the ONLY case where you announce before acting. For all other work, start immediately without status updates.
+
+### Background Task Policy:
+
+**Collect Strategist/Critic results before your final answer. No exceptions.**
+
+**Strategist/Critic-dependent implementation is BLOCKED until those results finish.**
+
+- If you asked Strategist/Critic for direction that affects the fix, do not implement before results arrive.
+- While waiting, only do non-overlapping prep work. Never ship implementation decisions they were asked to decide.
+- Never "time out and continue anyway" for Strategist/Critic-dependent tasks.
+
+- Strategist/Critic may take minutes. When done with your own work: **end your response** - wait for the \`<system-reminder>\`.
+- Do NOT poll \`background_output\` on a running Strategist/Critic task. The notification will come.
+- Never cancel Strategist/Critic consultation tasks.
+</strategist_critic_usage>`
+}
+
+export function buildNonClaudePlannerSection(model: string): string {
+  const isNonClaude = !model.toLowerCase().includes("claude")
+  if (!isNonClaude) {
+    return ""
+  }
+
+  return `### Strategist Dependency (Non-Claude)
+
+Multi-step task? **ALWAYS consult Strategist first.** Do NOT start implementation without a plan.
+
+- Single-file fix or trivial change → proceed directly
+- Anything else (2+ steps, unclear scope, architecture) → \`task(subagent_type="strategist", ...)\` FIRST
+- Use \`session_id\` to resume the same Strategist session - ask follow-up questions aggressively
+- If ANY part of the task is ambiguous, ask Strategist before guessing
+
+Strategist returns a structured work breakdown with parallel execution opportunities. Follow it.`
+}
+
+export function buildParallelDelegationSection(
+  model: string,
+  categories: AvailableCategory[],
+): string {
+  const isNonClaude = !model.toLowerCase().includes("claude")
+  const hasDelegationCategory = categories.some(
+    (category) => category.name === "deep" || category.name === "unspecified-high",
+  )
+
+  if (!isNonClaude || !hasDelegationCategory) {
+    return ""
+  }
+
+  return `### DECOMPOSE AND DELEGATE - YOU ARE NOT AN IMPLEMENTER
+
+**YOUR FAILURE MODE: You attempt to do work yourself instead of decomposing and delegating.** When you implement directly, the result is measurably worse than when specialized subagents do it. Subagents have domain-specific configurations, loaded skills, and tuned prompts that you lack.
+
+**for ANY implementation task:**
+
+1. **ALWAYS decompose** the task into independent work units. No exceptions. Even if the task "feels small", decompose it.
+2. **ALWAYS delegate** EACH unit to a \`deep\` or \`unspecified-high\` agent in parallel (\`run_in_background=true\`).
+3. **Keep execution contours separate**: bounded low-risk edits go to \`sub\`; deep, long-horizon implementation goes to \`coder\`.
+4. **NEVER work sequentially.** If 4 independent units exist, spawn 4 agents simultaneously. Not 1 at a time. Not 2 then 2.
+5. **NEVER implement directly** when delegation is possible. You write prompts, not code.
+
+**YOUR PROMPT TO EACH AGENT MUST INCLUDE:**
+- GOAL with explicit success criteria (what "done" looks like)
+- File paths and constraints (where to work, what not to touch)
+- Existing patterns to follow (reference specific files the agent should read)
+- Clear scope boundary (what is IN scope, what is OUT of scope)
+
+**Vague delegation = failed delegation.** If your prompt to the subagent is shorter than 5 lines, it is too vague.
+
+**Write code yourself** → Delegate to \`deep\` or \`unspecified-high\` agent
+**Handle 3 changes sequentially** → Spawn 3 agents in parallel
+**Send cheap bounded edits to coder** → Route those to \`sub\`
+**"Quickly fix this one thing"** → Still delegate - your "quick fix" is slower and worse than a subagent's
+
+**Your value is orchestration, decomposition, and quality control. Delegating with crystal-clear prompts IS your work.**`
+}