fix(forge): scope the primitive route to apm hooks

- say apm hook/instruction/prompt so pre-commit hooks stay with pc-author
- restore the idea-without-a-home trigger
- cite checkpoints forge's apm routes actually hit
- diff version bumps against the remote default branch

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KkT7RSDwDbmrM9T34b6sTi

plugins/kyberforge/.apm/skills/forge/SKILL.md

@@ -1,10 +1,10 @@
 ---
 name: forge
 description: >
-  Use when the user wants something built or improved but has not yet named
+  Use when the user wants to build or improve something of unnamed type ("where
-  its type ("skill or plugin?"). If named, use instead: skill -> `skill-author`,
+  does this idea go?"). If named, use instead: skill -> `skill-author`, agent ->
-  agent -> `agent-author`, hook/instruction/prompt -> `primitive-author`,
+  `agent-author`, apm hook/instruction/prompt -> `primitive-author`, plugin ->
-  plugin -> `apm-workflow`.
+  `apm-workflow`.
 metadata:
   version: "1.0.2"
   category: factory

plugins/kyberforge/.apm/skills/forge/references/apm-routes.md

@@ -15,15 +15,17 @@ removed per ADR-0015 once issue #90 landed, and `apm-workflow` is their sole suc
 ## Always inline, never forked
 
 Run these routes inline, in the current conversation. Their flows are short, prompt-heavy or
-gated — `apm publish` is not trivially reversible and removing a marketplace entry takes a
+gated — editing a package's manifest metadata republishes its public `plugin.json` description,
-conversational confirmation — and a backgrounded fork cannot surface those checkpoints to the
+and removing a marketplace entry takes a conversational confirmation before `apm pack` changes the
-user in real time.
+consumed catalog — and a backgrounded fork cannot surface those checkpoints to the user in real
+time.
 
 ## No clean-context recheck, and no automatic audit
 
-Skill, agent and primitive routes close with a clean-context audit rerun; these two do not, and the omission
+Skill, agent, and hook, instruction or prompt routes close with a clean-context audit rerun; these
-is deliberate rather than an oversight. Neither artifact type has an audit skill counterpart to
+two do not, and the omission is deliberate rather than an oversight. Neither artifact type has an
-re-run, so detaching the route to earn a recheck it would never get buys nothing.
+audit skill counterpart to re-run, so detaching the route to earn a recheck it would never get buys
+nothing.
 
 These routes get no automated terminal check either. `apm audit` is a separate action on
 `apm-workflow`'s own dispatch table, not a closing step of the configure or marketplace flow a

plugins/kyberforge/.apm/skills/forge/references/author-routes.md

@@ -8,8 +8,8 @@ source_keys:
 Reached from `SKILL.md` Step 2 when the classified artifact is a skill, an agent/subagent
 definition, or a hook, instruction or prompt. Route a skill to `skill-author`, an agent to
 `agent-author`, and a hook, instruction or prompt to `primitive-author`. The branches differ on
-the author skill only — all verify the result with `factory-audit`, which detects the artifact type
+the author skill only — all verify the result with `factory-audit`, which detects the artifact
-itself — and everything below applies to all of them.
+type itself — and everything below applies to all of them.
 
 ## Choose fork or inline
 
@@ -27,12 +27,12 @@ Fall back to an **inline invocation** — same conversation, no subagent — whe
 ## Two-tier verification
 
 Every author skill already closes out with its own inline audit, in the same context as the
-authoring work: `skill-author`, `agent-author` and `primitive-author` each invoke `factory-audit` on what they wrote.
+authoring work: `skill-author`, `agent-author` and `primitive-author` each invoke `factory-audit`
-That is tier one, and forge does not change it.
+on what they wrote. That is tier one, and forge does not change it.
 
 Tier two belongs to forge. Once the author skill's run has finished, spin up a separate
-**clean-context subagent** — fresh, not forked, no inherited context — to independently re-run the
+**clean-context subagent** — fresh, not forked, no inherited context — to independently re-run
-same audit skill against the finished artifact. This is a distinct verification layer, not a
+the same audit skill against the finished artifact. This is a distinct verification layer, not a
 duplicate: the inline audit shares context with the work it is checking and can share its blind
 spots, while the clean rerun has no stake in the result.
 

plugins/kyberforge/.apm/skills/forge/references/version-bump.md

@@ -25,9 +25,11 @@ Skip this step entirely if no ancestor `apm.yml` carries a `type:` field: the ar
 standalone or scoped to a user agent directory, and there is no package to version.
 
 Skip the bump, and report that you skipped it, if the branch already moved this package's `version`
-for unreleased work: `git diff $(git merge-base HEAD main) -- <package>/apm.yml` shows a changed
+for unreleased work: `git diff $(git merge-base HEAD origin/main) -- <package>/apm.yml` shows a
-`version:` line. One bump covers all unreleased work on a branch — `primitive-author` skips on the
+changed `version:` line. Diff against the remote default branch, not a local `main` that may be
-same condition — so bumping again here double-counts it.
+stale; if it is not `main`, resolve it with `git symbolic-ref refs/remotes/origin/HEAD`. One bump
+covers all unreleased work on a branch — `agent-author` and `primitive-author` skip on the same
+condition — so bumping again here double-counts it.
 
 ## Delegate the bump