chapterhouse 0.14.1 → 0.14.2

package/dist/copilot/agents.js

@@ -298,7 +298,7 @@ function getAgentBasePrompt() {
 You are an agent within Chapterhouse, a team-level AI assistant for engineering teams. You run on the user's local machine.
 
 ### Agent Memory
-Chapterhouse gives every agent a persistent, file-based memory tree under \`memory/\`, organized into domains. Read it with \`cog_read\`, \`cog_l0_scan\`, \`cog_l1_outline\`, \`cog_tree\`, and \`cog_search\`; write directly with \`cog_write\`, \`cog_edit\`, \`cog_append\`, and \`cog_move\` whenever you learn something worth remembering — don't wait for end of task. Every write auto-commits to git. The full memory operating instructions are included below.
+Every agent has a persistent memory store, organized into domains and reached **only** through the \`cog_*\` tools — never through \`bash\`, \`ls\`, \`find\`, or file reads, and you never need its filesystem path. Read with \`cog_l0_scan\`, \`cog_read\`, \`cog_search\`, \`cog_l1_outline\`, and \`cog_tree\`; write directly with \`cog_write\`, \`cog_edit\`, \`cog_append\`, and \`cog_move\` whenever you learn something worth remembering — don't wait for end of task. Every write auto-commits. The full memory operating instructions are included below.
 
 ### Shared Wiki
 All agents share a wiki knowledge base for persistent memory. Use \`wiki_read\` and \`wiki_search\` to find existing knowledge, and \`wiki_update\` to save important findings.

package/memory-assets/system-instructions.md

@@ -2,6 +2,12 @@
 
 Chapterhouse gives agents persistent memory, self-reflection, and foresight. It is the first layer of continuous awareness — not just recall, but cognition across time.
 
+## Accessing Memory
+
+Memory is reached **only** through the `cog_*` tools — `cog_l0_scan`, `cog_read`, `cog_l1_outline`, `cog_tree`, `cog_search`, `cog_stats`, `cog_write`, `cog_edit`, `cog_append`, `cog_move`, `cog_git`, `cog_domains`, `cog_domain_create`, `cog_sessions`. These are the complete and only interface.
+
+**Never** use `bash`, `ls`, `find`, `cat`, `grep`, or file-read tools to locate, inspect, or edit memory — they will not find it and you do not need them. Every `cog_*` path is relative to the memory root (e.g. `personal/hot-memory.md`); you never need, and must never search the filesystem for, an absolute memory path. Paths and directories named in this document are the `cog_*` path namespace, not shell paths. When you are unsure what exists, call `cog_l0_scan` first.
+
 ## Domain Routing & Skills
 
 Domains are defined in `memory/domains.yml`. Each domain skill and pipeline skill self-describes via its skill `description`, so the runtime routes to them automatically. Without a skill active, use judgment: casual conversation → personal domain; questions about a tracked domain → that domain.

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "chapterhouse",
-  "version": "0.14.1",
+  "version": "0.14.2",
   "description": "Chapterhouse — a team-level AI assistant for engineering teams, built on the GitHub Copilot SDK. Web UI only.",
   "bin": {
     "chapterhouse": "dist/cli.js"

package/skills/system/scenario/SKILL.md

@@ -2,6 +2,7 @@
 name: scenario
 description: Use for decision simulation — modeling 2-3 branches with timelines and contingencies. Trigger when the user says scenario, what if, simulate, model this, or play out.
 ---
+**Tool enforcement:** All reads and writes to `memory/` MUST use cog_* tools (`cog_read`, `cog_write`, `cog_edit`, `cog_append`, `cog_move`, `cog_tree`, `cog_l0_scan`, `cog_l1_outline`, `cog_search`, `cog_stats`, `cog_domains`, `cog_sessions`, `cog_git`). **Do NOT use bash, cat, echo, find, or any shell command to read or write memory files.** If a cog_* tool call fails, note the error in self-observations and fall back to bash only for that specific operation.
 
 Use this skill for scenario simulation — modeling decision branches with timelines, dependencies, and contingencies grounded in real memory data. Trigger if the user says "scenario", "what if", "model this", "simulate", "play out", "what happens if", or similar branching/decision-modeling requests. Also triggered when foresight flags a scenario candidate.
 

package/skills/system/setup/SKILL.md

@@ -2,6 +2,7 @@
 name: setup
 description: Use to bootstrap the memory system or add and reconfigure domains. Trigger when the user says setup, bootstrap, add a domain, or configure domains.
 ---
+**Tool enforcement:** All reads and writes to `memory/` MUST use cog_* tools (`cog_read`, `cog_write`, `cog_edit`, `cog_append`, `cog_move`, `cog_tree`, `cog_l0_scan`, `cog_l1_outline`, `cog_search`, `cog_stats`, `cog_domains`, `cog_domain_create`, `cog_sessions`, `cog_git`). **Do NOT use bash, cat, echo, find, or any shell command to read or write memory files.**
 
 This skill is **conversational** — you ask the user about their life and work, then generate `memory/domains.yml` and everything that flows from it. No one should ever need to manually edit `domains.yml`.