@zecalis/fael 0.41.0 → 0.43.0

package/README.md

@@ -5,12 +5,11 @@
 [![CI](https://github.com/zecalis/fael/actions/workflows/ci.yml/badge.svg)](https://github.com/zecalis/fael/actions/workflows/ci.yml)
 [![license](https://img.shields.io/badge/license-MIT-blue.svg)](LICENSE)
 
-**Your repo's experience, kept for every new agent.** Every agent arrives capable and new to your
-repo. fael keeps what the ones before it learned there — why a choice was made, what broke and how
-it was fixed, what is still open — in a shared work ledger. Hand work from one session, agent or
-person to the next — a decision, a requirement, an issue assigned to someone, a claim, where a plan
-stopped — and it shows up where the next agent is working: on the file it opens, and at session start
-for what is routed to it.
+**The ledger behind plan work.** Work runs as plans cut into chunks, one session or agent at a
+time, and no session shares another's chat. fael keeps what passes between them — where a chunk
+stopped, what was decided and why, the issue still open, what the user asked for — in a shared work
+ledger inside the repo, and hands it back when the next session picks the work up: at kickoff of the
+plan or file it names, on the file it opens, and at session start for what is routed to it.
 
 One person runs five agents; a team runs fifty — and each of those hands work to sub-agents. They
 don't share a chat, so every one starts from zero: it finds the same flaky test, re-asks why that
@@ -49,9 +48,10 @@ How it behaves:
   call. When a turn ends with no tool call left, it can close its reply with `fael
   decision|issue|note: … [files: …]` lines and fael files them — no turn is stopped or
   re-prompted.
-- **Memory finds them.** When an agent reads a file, the decisions and open bugs about *that file*
-  come attached — nobody has to remember to search. Sub-agents too: one starts with an empty
-  context and only a short brief, and the file it opens brings the memory the brief left out.
+- **Memory finds them.** When an agent edits a file, the decisions and open bugs about *that file*
+  come attached — nobody has to remember to search. Reading or searching brings nothing: the
+  rows wait for the edit. Sub-agents too: one starts with an empty context and only a short
+  brief, and the file it edits brings the memory the brief left out.
 - **No spam in context.** Rows are pushed per file, once per context window (a sub-agent, or a
   session after compaction, is told again — it no longer has them), and cut to a token budget
   (800 by default) — not a notes dump. Anything else the agent asks for itself, through MCP.

package/npm-shrinkwrap.json

@@ -19,7 +19,7 @@
       "hasInstallScript": true,
       "license": "MIT",
       "name": "@zecalis/fael",
-      "version": "0.41.0"
+      "version": "0.43.0"
     },
     "node_modules/detect-libc": {
       "engines": {
@@ -48,5 +48,5 @@
     }
   },
   "requires": true,
-  "version": "0.41.0"
+  "version": "0.43.0"
 }

package/package.json

@@ -1,6 +1,6 @@
 {
   "artifactDownloadUrls": [
-    "https://github.com/zecalis/fael/releases/download/v0.41.0"
+    "https://github.com/zecalis/fael/releases/download/v0.43.0"
   ],
   "bin": {
     "fael": "run-fael.js"
@@ -81,7 +81,7 @@
       "zipExt": ".tar.xz"
     }
   },
-  "version": "0.41.0",
+  "version": "0.43.0",
   "volta": {
     "node": "18.14.1",
     "npm": "9.5.0"