pluidr 0.2.0 → 0.3.1

package/src/templates/agent-prompts/writer.txt

@@ -1,18 +1,27 @@
 # Role: Writer Subagent
 
-You are a STATELESS FORMATTER. You transform structured input into a
-formatted document. You do not infer, decide, evaluate, or add content that
-wasn't given to you.
+You are the **Writer** subagent, the text-file executor for the **Builder**
+agent. You write and edit `.md` and `.txt` files across the project — like
+Coder but for documents only. You do not run bash commands.
 
-Refer to `hierarchy.txt` (loaded globally) — if the input you're given is
-incomplete, mark fields as `TBD`, do not invent content to fill gaps.
+## Mode: Document Creation / Editing
+
+You create and modify documentation files (.md, .txt) as instructed by
+Builder. You may also be invoked in Summary mode to produce completion
+reports from structured input.
 
 ## Delegation rules
 
 You have no `task` permission — you cannot invoke any other agent or
-subagent. If the input you receive is insufficient to produce a section,
-mark it `TBD` — do not attempt to research, infer, or ask another agent to
-fill the gap. That responsibility belongs to the caller (Planner/Builder/Reviewer).
+subagent. If Builder's instructions are insufficient to produce the
+required document, mark missing sections as `TBD` — do not infer, research,
+or invent content.
+
+## Principles
+
+- **Audience-First** — Structure the document for its reader. If Builder specifies an audience (technical lead, maintainer, end-user), tailor detail level and vocabulary accordingly. If no audience is specified, default to technical summary.
+- **DRY for Docs** — Each finding or verdict appears exactly once with cross-references. Do not repeat the same information across sections — it creates maintenance burden and risks inconsistency.
+- **KISS** — Prefer simple, flat structure over nested hierarchies. A reader should grasp the outcome from the first paragraph.
 
 ## Writer MUST NOT
 
@@ -20,43 +29,15 @@ fill the gap. That responsibility belongs to the caller (Planner/Builder/Reviewe
 - Make decisions (e.g., which approach is better).
 - Add analysis, recommendations, or opinions.
 - Fill missing information with assumptions — mark as `TBD` instead.
+- Run bash commands — you have no `bash` permission.
 
 ## Writer MAY ONLY
 
+- Write and edit `.md` and `.txt` files as instructed.
 - Reformat / restructure given input into the target document type.
-- Apply consistent terminology and structure (Principle of Least Astonishment —
-  same structure as prior documents of the same type).
-- Flag missing required fields explicitly (`TBD`) rather than silently
-  omitting or guessing them.
-
-## Modes (caller specifies which one)
-
-### Mode: PRD (called by Planner)
-Input: goal, requirements, assumptions, open questions (resolved).
-Output: structured PRD document, saved to `docs/plans/<slug>.md`.
-
-```markdown
-# PRD: <title>
-
-## Goal
-<one sentence, as given>
-
-## Requirements
-- R1: <as given>
-- R2: ...
-
-## Assumptions
-- <as given, or "None">
-
-## Out of Scope
-- <as given, or "Not specified — TBD">
-```
+- Flag missing required fields explicitly (`TBD`).
 
-### Mode: Summary/Report (called by Builder or Reviewer)
-Input: Verifier output, Debugger RCA, or task completion notes.
-Output: audience-appropriate summary, saved to `docs/reports/<slug>.md` —
-no new analysis, just compression and structure (DRY for docs: don't repeat
-the same point across sections).
+## Output Format (Summary/Report mode)
 
 ```markdown
 # Report: <subject>
@@ -70,5 +51,5 @@ the same point across sections).
 <compressed from input, no added interpretation>
 ```
 
-If the input to any mode doesn't give you enough to fill a section, write
-`TBD` and move on. Never write "I think..." or "this probably means...".
+If the input to any section doesn't give you enough to fill it, write `TBD`
+and move on. Never write "I think..." or "this probably means...".

package/src/templates/model-defaults.json

@@ -2,11 +2,11 @@
   "reasoningHeavy": {
     "provider": "opencode",
     "model": "big-pickle",
-    "agents": ["planner", "reviewer", "debugger", "verifier"]
+    "agents": ["planner", "debugger", "explorer", "researcher", "plan-writer", "plan-checker", "inspector", "reporter"]
   },
   "fast": {
     "provider": "opencode",
     "model": "deepseek-v4-flash-free",
-    "agents": ["builder", "researcher", "coder", "writer"]
+    "agents": ["builder", "coder", "tester", "reviewer", "writer", "fixer"]
   }
 }

package/src/templates/opencode.config.json

@@ -1,6 +1,6 @@
 {
   "$schema": "https://opencode.ai/config.json",
-  "instructions": ["prompts/hierarchy.txt"],
+  "instructions": ["{file:./prompts/hierarchy.txt}"],
   "permission": {
     "read": {
       "*.env": "deny",
@@ -12,31 +12,62 @@
     "plan": { "hidden": true },
     "build": { "hidden": true },
 
+    "explorer": {
+      "mode": "primary",
+      "temperature": 0.7,
+      "description": "Brainstorms with user, scans codebase + web, recommends to Planner. No subagents — research and recommendations only.",
+      "prompt": "{file:./prompts/explorer.txt}",
+      "permission": {
+        "read": "allow",
+        "glob": "allow",
+        "grep": "allow",
+        "edit": "deny",
+        "write": "deny",
+        "webfetch": "allow",
+        "websearch": "allow",
+        "question": "allow",
+        "bash": {
+          "*": "deny",
+          "git log*": "allow",
+          "git diff*": "allow",
+          "git show*": "allow",
+          "git status*": "allow",
+          "rg *": "allow",
+          "grep *": "allow"
+        },
+        "task": "deny"
+      }
+    },
+
     "planner": {
       "mode": "primary",
-      "description": "Turns a user request into a verified PRD via Researcher/Writer/Verifier, then decides build-or-revise. Cannot write or edit code.",
+      "description": "Turns a user request into a verified PRD via researcher/plan-writer/plan-checker, then decides build-or-revise. Cannot write or edit code.",
       "prompt": "{file:./prompts/planner.txt}",
       "permission": {
+        "read": "allow",
+        "glob": "allow",
+        "grep": "allow",
         "edit": "deny",
         "write": "deny",
         "bash": "deny",
-        "webfetch": "allow",
-        "websearch": "allow",
         "question": "allow",
         "task": {
           "*": "deny",
           "researcher": "allow",
-          "writer": "allow",
-          "verifier": "allow"
+          "plan-writer": "allow",
+          "plan-checker": "allow"
         }
       }
     },
 
     "builder": {
       "mode": "primary",
-      "description": "Executes a confirmed PRD by orchestrating Coder/Tester/Verifier/Writer. Cannot edit files, run bash, or change requirements directly.",
+      "description": "Executes a confirmed PRD by orchestrating coder/tester/reviewer/writer. Cannot edit files, run bash, or change requirements directly.",
       "prompt": "{file:./prompts/builder.txt}",
       "permission": {
+        "read": "allow",
+        "glob": "allow",
+        "grep": "allow",
         "edit": "deny",
         "write": "deny",
         "bash": "deny",
@@ -45,79 +76,70 @@
           "*": "deny",
           "coder": "allow",
           "tester": "allow",
-          "verifier": "allow",
+          "reviewer": "allow",
           "writer": "allow"
         }
       }
     },
 
-    "reviewer": {
+    "debugger": {
       "mode": "primary",
-      "temperature": 0.1,
-      "description": "Reviews completed work for traceability and regression risk via Verifier/Debugger/Researcher/Writer. Cannot edit or write code.",
-      "prompt": "{file:./prompts/reviewer.txt}",
+      "description": "Root-cause analysis via inspector/fixer/reporter. Standalone primary agent — does not depend on Builder. Cannot edit files or run bash directly.",
+      "prompt": "{file:./prompts/debugger.txt}",
       "permission": {
         "read": "allow",
         "glob": "allow",
         "grep": "allow",
         "edit": "deny",
         "write": "deny",
-        "bash": {
-          "*": "deny",
-          "git diff*": "allow",
-          "git log*": "allow",
-          "git show*": "allow",
-          "git blame*": "allow",
-          "rg *": "allow"
-        },
+        "bash": "deny",
+        "question": "allow",
         "task": {
           "*": "deny",
-          "verifier": "allow",
-          "debugger": "allow",
-          "researcher": "allow",
-          "writer": "allow"
+          "inspector": "allow",
+          "fixer": "allow",
+          "reporter": "allow"
         }
       }
     },
 
-    "researcher": {
+    "coder": {
       "mode": "subagent",
-      "description": "Researches technical context (library/API/best practice) and existing codebase patterns. Outputs confirmed_facts / inferred_facts / unknowns / risks only — never blended.",
-      "prompt": "{file:./prompts/researcher.txt}",
+      "description": "Implements tasks assigned by Builder, following the PRD exactly. Tracks own progress via todowrite.",
+      "prompt": "{file:./prompts/coder.txt}",
       "permission": {
         "read": "allow",
         "glob": "allow",
         "grep": "allow",
-        "edit": "deny",
-        "write": "deny",
-        "bash": "deny",
-        "webfetch": "allow",
-        "websearch": "allow",
+        "edit": "allow",
+        "write": "allow",
+        "todowrite": "allow",
+        "bash": "allow",
         "task": "deny"
       }
     },
 
-    "coder": {
+    "tester": {
       "mode": "subagent",
-      "description": "Implements tasks assigned by Builder, following the PRD exactly. Tracks own progress via todowrite.",
-      "prompt": "{file:./prompts/coder.txt}",
+      "description": "Runs tests on implemented code and reports results (PASS/FAIL/BLOCKED) with coverage gaps. Cannot fix code, redesign tests, or decide next steps.",
+      "prompt": "{file:./prompts/tester.txt}",
       "permission": {
         "read": "allow",
         "glob": "allow",
         "grep": "allow",
-        "edit": "allow",
-        "write": "allow",
-        "todowrite": "allow",
+        "edit": "deny",
+        "write": "deny",
+        "todowrite": "deny",
         "bash": "allow",
         "task": "deny"
       }
     },
 
-    "debugger": {
+    "reviewer": {
       "mode": "subagent",
       "temperature": 0.1,
-      "description": "Diagnoses defects via reproduce -> isolate -> source -> remedy, called by Reviewer. Cannot fix code directly.",
-      "prompt": "{file:./prompts/debugger.txt}",
+      "description": "Boolean gate for Builder: compares implementation against PRD definition-of-done. PASS/FAIL + gap list only.",
+      "prompt": "{file:./prompts/reviewer.txt}",
       "permission": {
         "read": "allow",
         "glob": "allow",
@@ -128,36 +150,92 @@
           "*": "deny",
           "git diff*": "allow",
           "git log*": "allow",
+          "git show*": "allow",
           "rg *": "allow",
-          "grep *": "allow",
-          "find *": "allow",
-          "cat *": "allow"
+          "grep *": "allow"
         },
         "task": "deny"
       }
     },
 
-    "tester": {
+    "writer": {
       "mode": "subagent",
-      "description": "Runs tests on implemented code and reports results (PASS/FAIL/BLOCKED) with coverage gaps. Cannot fix code, redesign tests, or decide next steps.",
-      "prompt": "{file:./prompts/tester.txt}",
+      "description": "Text-file executor for Builder. Writes and edits .md/.txt files across the project — like coder but for documents only. No bash.",
+      "prompt": "{file:./prompts/writer.txt}",
+      "permission": {
+        "read": "allow",
+        "glob": "allow",
+        "grep": "allow",
+        "edit": "allow",
+        "write": "allow",
+        "bash": "deny",
+        "todowrite": "deny",
+        "task": "deny"
+      }
+    },
+
+    "researcher": {
+      "mode": "subagent",
+      "description": "Researches technical and codebase context for Planner during PRD preparation. Outputs confirmed_facts / inferred_facts / unknowns / risks only.",
+      "prompt": "{file:./prompts/researcher.txt}",
       "permission": {
         "read": "allow",
         "glob": "allow",
         "grep": "allow",
         "edit": "deny",
         "write": "deny",
-        "todowrite": "deny",
-        "bash": "allow",
+        "bash": "deny",
+        "webfetch": "allow",
+        "websearch": "allow",
+        "task": "deny"
+      }
+    },
+
+    "plan-writer": {
+      "mode": "subagent",
+      "description": "Stateless formatter for Planner. Produces PRD documents from structured input — no inference, no decisions. PRD mode only.",
+      "prompt": "{file:./prompts/plan-writer.txt}",
+      "permission": {
+        "read": "allow",
+        "glob": "allow",
+        "grep": "allow",
+        "edit": {
+          "*": "deny",
+          "docs/plans/**": "allow"
+        },
+        "write": {
+          "*": "deny",
+          "docs/plans/**": "allow"
+        },
+        "bash": "deny",
         "task": "deny"
       }
     },
 
-    "verifier": {
+    "plan-checker": {
       "mode": "subagent",
       "temperature": 0.1,
-      "description": "Boolean gate: compares two artifacts (PRD vs request, code vs DoD, PRD vs code) and reports PASS/FAIL with a gap list. Never proposes features or redesigns.",
-      "prompt": "{file:./prompts/verifier.txt}",
+      "description": "Boolean gate for Planner: checks PRD completeness against original user request. PASS/FAIL + gap list only.",
+      "prompt": "{file:./prompts/plan-checker.txt}",
+      "permission": {
+        "read": "allow",
+        "glob": "allow",
+        "grep": "allow",
+        "edit": "deny",
+        "write": "deny",
+        "bash": {
+          "*": "deny",
+          "rg *": "allow",
+          "grep *": "allow"
+        },
+        "task": "deny"
+      }
+    },
+
+    "inspector": {
+      "mode": "subagent",
+      "description": "Root-cause analysis for Debugger. Investigates bugs via codebase and web research. Outputs confirmed_facts / inferred_facts / unknowns / risks only.",
+      "prompt": "{file:./prompts/inspector.txt}",
       "permission": {
         "read": "allow",
         "glob": "allow",
@@ -169,28 +247,48 @@
           "git diff*": "allow",
           "git log*": "allow",
           "git show*": "allow",
+          "git blame*": "allow",
           "rg *": "allow",
-          "grep *": "allow"
+          "grep *": "allow",
+          "cat *": "allow",
+          "find *": "allow"
         },
+        "webfetch": "allow",
+        "websearch": "allow",
         "task": "deny"
       }
     },
 
-    "writer": {
+    "fixer": {
       "mode": "subagent",
-      "description": "Stateless formatter. Produces PRD documents (docs/plans/) and Summary/Report documents (docs/reports/) from structured input only — no inference, no decisions.",
-      "prompt": "{file:./prompts/writer.txt}",
+      "description": "Implements bug fixes for Debugger. Follows root cause from inspector. Like coder but focused on fixes — minimal change, root-cause targeting.",
+      "prompt": "{file:./prompts/fixer.txt}",
       "permission": {
         "read": "allow",
         "glob": "allow",
+        "grep": "allow",
+        "edit": "allow",
+        "write": "allow",
+        "todowrite": "allow",
+        "bash": "allow",
+        "task": "deny"
+      }
+    },
+
+    "reporter": {
+      "mode": "subagent",
+      "description": "Stateless formatter for Debugger. Produces diagnosis + remedy reports from structured input — no inference, no decisions. Summary mode only.",
+      "prompt": "{file:./prompts/reporter.txt}",
+      "permission": {
+        "read": "allow",
+        "glob": "allow",
+        "grep": "allow",
         "edit": {
           "*": "deny",
-          "docs/plans/**": "allow",
           "docs/reports/**": "allow"
         },
         "write": {
           "*": "deny",
-          "docs/plans/**": "allow",
           "docs/reports/**": "allow"
         },
         "bash": "deny",

package/src/templates/agent-prompts/verifier.txt

@@ -1,68 +0,0 @@
-# Role: Verifier Subagent
-
-You are a GATE, not a reasoning layer. You compare two artifacts and report
-PASS/FAIL with a structured gap list. You do not design, suggest improvements,
-or make decisions beyond the comparison itself.
-
-Refer to `hierarchy.txt` (loaded globally) — you do not resolve conflicts
-yourself; you report them as gaps.
-
-## Delegation rules
-
-You have no `task` permission — you cannot invoke any other agent or
-subagent, and you cannot ask Researcher to fill in missing context. If you
-lack enough information to render a verdict, say so as a gap ("cannot
-verify X — insufficient input") rather than guessing or treating silence as PASS.
-
-## Verifier MUST NOT
-
-- Propose new features or scope.
-- Redesign architecture or suggest alternative approaches.
-- Suggest improvements beyond what's needed to close a gap against the
-  artifact you're comparing against.
-- Make a "lanjut atau revisi" decision — that belongs to Planner/Builder/Reviewer.
-
-## Verifier MAY ONLY output
-
-- PASS / FAIL verdict
-- Gap list (requirement/expectation ↔ what's actually there)
-- Requirement → artifact mapping (traceability)
-
-## Modes (context determines which one applies — caller specifies)
-
-### Mode Planner: Check PRD (called by Planner)
-Compare PRD draft against the original user request / internal MoM notes.
-- Focus: completeness (every request element has a corresponding requirement),
-  ambiguity (any requirement that could be read 2+ ways), contradiction
-  (requirements that conflict with each other).
-- Output: PASS, or FAIL + list of {missing | ambiguous | contradictory} items.
-
-### Mode Builder: Check Implementation (called by Builder)
-Compare implementation against PRD's definition-of-done for each task.
-- Focus: does the code satisfy the stated DoD — nothing more, nothing less.
-- Output: PASS, or FAIL + list of {task_id, expected, actual, gap}.
-
-### Mode Reviewer: Compare PRD vs Code (called by Reviewer)
-Full traceability pass after implementation is "done."
-- Focus: every PRD requirement maps to specific code; flag any code that
-  doesn't trace back to a requirement (scope creep) and any requirement
-  with no corresponding code (gap).
-- Output: PASS, or FAIL + {requirement → code mapping table} + {orphan code}
-  + {unmet requirements}.
-
-## Output Format (all modes)
-
-```markdown
-## Verdict: PASS | FAIL
-
-## Mapping
-| Requirement/Expectation | Found in artifact? | Note |
-|---|---|---|
-
-## Gaps (if FAIL)
-- <item>: <expected> vs <actual>
-```
-
-If you find yourself wanting to write "I suggest..." or "it would be better
-to...", stop — that is out of scope. Report it as a gap instead and let the
-calling agent decide what to do about it.