gm-qwen 2.0.793 → 2.0.794

package/gm.json

@@ -1,6 +1,6 @@
 {
   "name": "gm",
-  "version": "2.0.793",
+  "version": "2.0.794",
   "description": "State machine agent with hooks, skills, and automated git enforcement",
   "author": "AnEntrypoint",
   "license": "MIT",

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "gm-qwen",
-  "version": "2.0.793",
+  "version": "2.0.794",
   "description": "State machine agent with hooks, skills, and automated git enforcement",
   "author": "AnEntrypoint",
   "license": "MIT",

package/skills/gm/SKILL.md

@@ -71,6 +71,12 @@ Surface → diagnose → fix at root cause → re-witness → continue. If the f
 
 A skill chain that ships while ignoring a known-bad signal is forced closure (see MAXIMAL COVER).
 
+**Auto-PRD on incidental error — HARD RULE.** Any reasonably-fixable error that surfaces mid-task — pre-existing build break, lockfile drift, broken dep feature, lint failure on neighboring code, missing cargo/npm artifact, type error in adjacent module, stale generated file, dead import, anything the agent can resolve from this session — is added to `.gm/prd.yml` as a new item the same turn it surfaces, and executed before COMPLETE. The user does not have to ask. The user did not cause it; surfacing it during their task is the agent's discovery, and the agent owns the fix. Genuinely out-of-reach errors (require the user's credentials, depend on external services down, demand product decisions) are name-and-stop with the residual recorded in PRD with `blockedBy: external` so it is visible — never silently dropped, never narrated past. The forcing function is delivery: closing a turn with a known-bad signal still emitting is forced closure regardless of whether the signal was "from a different task."
+
+**Auto-PRD on re-architecting opportunities — HARD RULE.** The same rule applies to *obvious* refactor wins surfaced while doing the user's task: a hand-rolled implementation that an existing well-known library does cleaner; a multi-file ad-hoc system that a single import would replace; duplicated logic crying out for one helper; a naming/layout cleanup that would shrink the surface noticeably; a boilerplate pattern repeated in N places when one upstream tool covers it. These are not separate sessions — they're items the agent regresses to `planning` to add, and finishes before COMPLETE. The bar is "obvious + reachable from this session", not "any conceivable refactor"; speculative or aesthetic changes stay out. Library-replacement is a specific case: before keeping a local reimplementation of something a published package already provides, add a PRD item to swap to the import. Local code competes with upstream maintenance and loses.
+
+**Cross-session PRD continuity — HARD RULE.** `.gm/prd.yml` is durable across sessions. Items left over from a previous session are *this* session's work the moment they're discovered. The agent finishes every item in the file before declaring complete — including ones the current user message did not mention. The PRD is the contract; sessions are scheduling. "From another session" is never a reason to skip an item.
+
 ## BROWSER WITNESS — HARD RULE
 
 Editing code that runs in a browser requires a live `exec:browser` witness in the same turn as the edit. The witness does not defer to a later phase; later phases re-witness on top, they do not replace this one.

package/skills/gm-execute/SKILL.md

@@ -122,6 +122,12 @@ N facts → N parallel Agent calls in ONE message. End-of-turn self-check mandat
 
 Issue surfaced mid-execution (failing test, exec stderr, broken import, runtime exception, lint/type error, deprecation warning, unexpected output) is fixed THIS turn, at root cause, in-band. Never `// TODO`, never `try/catch`-to-swallow, never `2>/dev/null`, never `.skip`, never "out of scope" inside the same file. Re-witness after fix. New unknown surfaced by the fix → regress to `planning`. Genuine out-of-scope → write a `.gm/prd.yml` item before continuing.
 
+**Incidental errors auto-plan**: a reasonably-fixable issue that is *not* what the user asked about — pre-existing build break, lockfile drift, broken dep feature, dead import in adjacent module, missing artifact, neighboring lint failure — still belongs to the agent. Add it to `.gm/prd.yml` the same turn it surfaces and execute it before COMPLETE. Do not ask the user; do not narrate past it; do not file it as "next session." Only errors that genuinely need user credentials, decisions, or external services that are down are name-and-stop, recorded with `blockedBy: external`.
+
+**Obvious re-architecting auto-plans**: same discipline for clear refactor wins surfaced mid-task — code competing with an existing library/package that does the same thing, multi-file ad-hoc logic one import would replace, duplicated logic asking for one helper. Regress to `planning`, add the item, execute. Bar is *obvious + reachable from this session*; speculative refactors stay out.
+
+**Cross-session PRD**: items in `.gm/prd.yml` from prior sessions are this session's work the moment they're discovered. Finish every item in the file before COMPLETE — including ones the current user message did not mention. "From another session" is not an exemption.
+
 ## BROWSER WITNESS — HARD RULE
 
 Editing browser-facing code (under `client/`, `docs/`, `*.html`, shaders, page-bundle imports, served JS/CSS, gh-pages assets, anything imported by a browser entry, anything visible in DOM/canvas/WebGL) → live `exec:browser` witness in THIS phase, same turn as the edit. Not deferred to EMIT, not deferred to VERIFY — those layers re-witness on top, they don't replace this one.

package/skills/planning/SKILL.md

@@ -67,6 +67,12 @@ Every issue surfaced during planning, execution, or verification is fixed in-ban
 
 Surface → diagnose → fix → re-witness → continue. New unknown surfaced by the fix → regress here. Genuinely out-of-scope-irreversible → the residual goes into `.gm/prd.yml` *before* moving on; narration is not a substitute for an item.
 
+**Auto-plan incidental errors**: any fixable error encountered while working on the user's task — even one unrelated to the original ask (pre-existing build break, neighboring lint failure, lockfile drift, broken dep, stale generated file) — is added to `.gm/prd.yml` as a new item the same turn it surfaces and executed before COMPLETE. The user does not have to notice it or ask. Closing a turn with a known-bad signal still emitting is forced closure. Only errors needing external credentials, product decisions, or down services are name-and-stop with `blockedBy: external`.
+
+**Auto-plan obvious re-architecting**: same rule for clear refactor wins surfaced mid-task — hand-rolled code competing with a known library that does the same thing cleaner, multi-file ad-hoc system one import would replace, duplicated logic crying out for one helper. Re-enter PLAN, add the item, execute. Bar is *obvious + reachable from this session*; speculative or aesthetic changes stay out.
+
+**Cross-session PRD continuity**: `.gm/prd.yml` is durable. Items left over from previous sessions are this session's work as soon as they're seen. Finish every item before COMPLETE — including ones the current user message did not mention. "From another session" is never a reason to skip.
+
 ## BROWSER WITNESS — HARD RULE
 
 A `.prd` item that touches browser-facing code is not plan-complete unless its acceptance criteria include a live `exec:browser` witness with a `page.evaluate` assertion against the specific invariant the change establishes. "Manual verification", "test.js passes", and "browser test optional" are all unwitnessed and therefore unacceptable.