@plurnk/plurnk-grammar 0.72.0 → 0.73.1

package/dist/plurnk.gbnf

@@ -302,7 +302,10 @@ status-mid ::= "10" [0-13-9] | "1" [1-9] [0-9] | "20" [13-9] | "2" [1-9] [0-9] |
 tags ::= "[" tag tag-rest* "]"
 tag ::= [A-Za-z0-9_.-]+
 tag-rest ::= "," tag
-target ::= "(" [^\x00-\x08\x0B-\x0C\x0E-\x1F\x7F)<\r\n]+ ")"
+target ::= "(" target-inner ")" | "(" target-regex ")"
+target-inner ::= [^\x00-\x08\x0B-\x0C\x0E-\x1F\x7F)<\r\n]+
+target-regex ::= "#" target-rx-char* "#" [a-zA-Z]*
+target-rx-char ::= "\\" [^\r\n] | [^#\r\n]
 line ::= "<" int line-rest* ">"
 line-rest ::= "," " "? int
 int ::= "-"? [0-9]+ frac?

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@plurnk/plurnk-grammar",
-  "version": "0.72.0",
+  "version": "0.73.1",
   "description": "ANTLR4 grammar for the Plurnk LLM agent protocol",
   "type": "module",
   "license": "MIT",

package/plurnk.md

@@ -20,13 +20,13 @@ Slots between `<<OPsuffix` and `:body:` are all optional. `:body:` fences are re
 | MOVE | apply tags  | required   | lines `N,M`         | destination URI  |
 | OPEN | filter tags | log path   | results `N,M`       | matcher          |
 | FOLD | filter tags | log path   | results `N,M`       | matcher          |
-| KILL | —           | required   | —                   | —                |
+| KILL | signal      | required   | —                   | —                |
 | EXEC | executor    | cwd        | —                   | command or code  |
 | SEND | status code | recipient  | —                   | message body     |
 
 Operations emit their status and/or results on the subsequent turn.
 READ output prefixes every line with line numbers and a hard tab, `N:	`. The prefix is not part of the source.
-EDIT is only for entries. Do not attempt to edit log items.
+EDIT is only for adding or modifying entries. Do not attempt to edit log items.
 EXEC defaults to `sh`; override with an optional executor (`sqlite`, `node`, etc.).
 
 ## Context
@@ -36,7 +36,7 @@ The agent maintains two surfaces for budgeting working-memory tokens:
 - **Log** — the record of every operation. FOLD contracts a log row to its one-line summary and saves tokens; OPEN shows the complete record but spends from your `tokensFree` context tokens. Non-destructive — FOLDed rows remain listed and re-OPENable.
 - **`plurnk:///manifest.json`** — what's available: the complete unranked directory of every entry. Query it to discover available entries.
 
-OPEN and FOLD operate on the log only. Log items are read-only, but can be KILLed.
+OPEN and FOLD operate on the log only. Log items are read-only, but can be KILLed (erased).
 
 ## `<Line> / <Result>`
 
@@ -77,12 +77,12 @@ URI-shaped: `[scheme://]rest`.
 * Bare paths (no scheme) default to local relative project file paths.
 * Glob metacharacters match within path segments; a standalone `#pattern#flags` matches the whole target by regex.
 * Path suffix (`.json`, `.md`, `.txt`, etc.) declares mimetype.
-* A literal `)` closes the target — percent-encode parens in a path (`%28`/`%29`).
+* Percent-encode reserved characters in paths: `)`→`%29`, `<`→`%3C`.
 * Append `#channel` to select a channel (e.g. `#stdout`, `#stderr`); absent, the scheme's default channel is used.
 
 ## Suffix
 
-When quoting plurnk operations in a body, YOU MUST use a matching single-digit suffix (`1`–`9`) on the opening and closing tags.
+When quoting plurnk operations in a body, YOU MUST use a matching single-digit suffix (`1`–`9`) or label (`[a-z]+`) on the opening and closing tags.
 
 <<EDIT1(known:///demo):
 quoted: <<EDIT(known:///inner):hello:EDIT
@@ -120,7 +120,7 @@ Body content is character-perfect, exactly matching whitespace.
 <<COPY[archive,2026-05-14](known:///draft.md):known:///archive/2026-05-14/draft.md:COPY
 <<MOVE[final](known:///draft/answer.md):known:///final/answer.md:MOVE
 <<KILL(obsolete/file.md)::KILL
-<<KILL(sh:///3/1/2)::KILL
+<<KILL[9](sh:///3/1/2)::KILL
 <<EDIT(run://capital-checker):Find the capital of France.:EDIT
 <<COPY(run://.):Re-derive the capital from a primary source.:COPY
 <<OPEN(log:///**/get)<1,10>::OPEN
@@ -129,7 +129,7 @@ Body content is character-perfect, exactly matching whitespace.
 <<FOLD(log:///**/get)<101,200>::FOLD
 <<FIND(log:///**/error):#timeout|deadline exceeded#i:FIND
 <<FIND(known:///**):revolution:FIND
-<<FIND(#draft.*#i)::FIND
+<<FIND(#(draft|final)/.*#i)::FIND
 <<FIND(#src/.*\.test\.ts#)::FIND
 <<FIND(src/**):@<createCoder:FIND
 
@@ -159,6 +159,6 @@ YOU SHOULD leverage taxonomic path names, folksonomic tags, and bulk pattern ope
 YOU MUST use OPEN and FOLD to keep your context budget healthy, optimized, topical, and below the `tokensFree` limit.
 YOU MUST terminate the turn by SENDing a status code containing the results, answer, or a status update: `<<SEND[N]:...:SEND`
 YOU MUST terminate a continuing loop with status code 102: <<SEND[102]:Forking a research run, optimizing log relevance...:SEND
-YOU MUST terminate a final turn with status code 200: <<SEND[200]:Paris:SEND
-YOU MUST terminate an idle/waiting loop with status code 202: <<SEND[202]:Parked until the capital-checker reports.:SEND
 YOU MUST terminate a failed/aborted loop with status code 499: <<SEND[499]:Giving up — cannot identify the capital from available sources.:SEND
+YOU MUST terminate a final turn with status code 200: <<SEND[200]:Paris:SEND
+YOU MUST pause an idle/waiting loop with status code 202: <<SEND[202]:Waiting until the capital-checker reports.:SEND