single-source installs: retire the dist-publish hop (install from iap/pstack-hermes/pstack)

.gitattributes

@@ -6,3 +6,7 @@
 # Mark it generated so GitHub collapses its diffs and excludes it from
 # language stats: review it as build output, not as hand-written code.
 pstack/** linguist-generated=true
+
+# Binaries stay byte-identical on checkout (assets/logo.png): pin the intent
+# so a future text rule cannot start rewriting them.
+*.png binary

.github/actions/build-package/action.yml

@@ -2,16 +2,16 @@ name: Build pstack package
 description: >-
   Shared build prologue for every workflow that produces the generated pstack/
   package: pin the upstream clone, verify it resolved, convert, validate and
-  scanner-gate. Extracted from the step bodies that ci.yml, release.yml and
-  publish-plugin.yml each carried their own copy of, so all three callers now
-  get the same fail-loud SHA verification.
+  scanner-gate. Extracted from the step bodies that ci.yml and release.yml each
+  carried their own copy of, so every caller now gets the same fail-loud SHA
+  verification.
 
 # WHY A COMPOSITE ACTION AND NOT A REUSABLE WORKFLOW
 # A reusable workflow (`uses: ./.github/workflows/build.yml`) would have to contain
 # the caller's ENTIRE job, because a `uses:` job cannot be followed by more steps
-# in that same job. All three callers continue past the shared prologue with their
+# in that same job. Both callers continue past the shared prologue with their
 # own steps (ci: reproducibility gate + artifact upload; release: zip + notes +
-# gh release; publish: dist push + tag), so a reusable workflow cannot express
+# gh release), so a reusable workflow cannot express
 # this shape without duplicating those steps into the called file. A composite
 # action composes into the middle of a job and is the correct tool here.
 # The usual reason to prefer a reusable workflow - secrets - does not apply: this

.github/workflows/ci.yml

@@ -49,8 +49,8 @@ jobs:
       # job above validates the source pstack/ directory on disk, where files are LF
       # because the converter wrote them that way. Nothing here constructs a git
       # checkout of the dist tree, so a line-ending or binary-asset regression in the
-      # published artifacts would ship uncaught. This stages the package exactly as
-      # publish-plugin.yml does and checks out the result with core.autocrlf=true.
+      # published artifacts would ship uncaught. This stages the package exactly as an
+      # install clone receives it (pstack/ subdir, core.autocrlf=true).
       - name: Dist checkout is LF-only under autocrlf
         shell: bash
         run: uv run --frozen tools/dist_checkout_check.py --package pstack --source-tree

.github/workflows/release.yml

@@ -28,7 +28,7 @@ jobs:
           persist-credentials: false
 
       # release.yml previously carried its own copy of the build prologue AND was
-      # missing the "Verify pinned SHA" step that ci.yml and publish-plugin.yml
+      # missing the "Verify pinned SHA" step that ci.yml
       # had, so a release could be cut from a pin that silently failed to
       # resolve. The shared action gives every caller the verified-pin behaviour.
       - name: Build package from pin

.github/workflows/upstream-drift-watch.yml

@@ -92,8 +92,8 @@ jobs:
             echo ""
             echo "Rebuild against the new upstream, update the transform maps for any"
             echo "reworded anchors (the converter's anchor audit will name them), bump"
-            echo "UPSTREAM_PIN in all four workflows (ci.yml / release.yml /"
-            echo "publish-plugin.yml / upstream-drift-watch.yml - tools/check_pins.py"
+            echo "UPSTREAM_PIN in all three workflows (ci.yml / release.yml /"
+            echo "upstream-drift-watch.yml - tools/check_pins.py"
             echo "enforces lock-step), and re-run the pipeline."
           } > "$RUNNER_TEMP/drift-body.md"
           # Guard for the next step: split across steps, the no-drift `exit 0`s

AGENTS.md

@@ -38,9 +38,9 @@ to improve **stock** pstack skills shipped in this package.
 |---|---|---|
 | Change stock method for all users | Converter / `tools/assets/` → rebuild → validate → release | Yes (product) |
 | Personal or project procedure | `skill_manage` / `/learn` into **local** `~/.hermes/skills/` | Yes (outside package) |
-| Owned variant of pstack skills | **Fork** of dist `iap/pstack`, install from the fork | Yes (you control merge) |
+| Owned variant of pstack skills | **Fork** of this repo, install from the fork's `pstack/` subdir | Yes (you control merge) |
 | New skills beside pstack | **Skill tap** (`hermes skills tap add` / install) | Yes |
-| Durable "learning" into installed plugin `SKILL.md` files | **Forbidden** as source of truth | No — updates and dist republish collide |
+| Durable "learning" into installed plugin `SKILL.md` files | **Forbidden** as source of truth | No — reinstalls overwrite the installed tree |
 
 > [!WARNING]
 > Never instruct `skill_manage` (or any self-evolution path) to patch
@@ -75,7 +75,7 @@ Default home for work is **this repo**. Escalate only when the table says so.
 | If you need to… | Do this | Not this |
 |---|---|---|
 | Change skill prose, delegation wording, model panel, hermesbot, bans, docs, converter maps | Edit `tools/` / `tools/assets/` → rebuild → validate → (ledger row if converter changed) | Hand-edit `pstack/`; rewrite in Cursor vocabulary |
-| Personal skill improvement or new procedures | Local skills, skill tap, or fork of `iap/pstack` ([docs/USING.md](docs/USING.md)) | `skill_manage` against stock plugin skills; durable edits under the install tree |
+| Personal skill improvement or new procedures | Local skills, skill tap, or fork of this repo ([docs/USING.md](docs/USING.md)) | `skill_manage` against stock plugin skills; durable edits under the install tree |
 | Package validates / doctor is clean, but stock install misbehaves on hermes | File/fix in `NousResearch/hermes-agent`; `patches/` is **historical reference only** ([docs/PATCHES.md](docs/PATCHES.md)) | Paper over harness gaps with skill prose that claims nonexistent tools |
 | Bug reproduces on upstream Cursor pstack *without* this port | Route to `cursor/plugins` ([SECURITY.md](SECURITY.md)) | "Fix" it in the converter as if it were a port bug |
 | Touch `.cursor-plugin/`, `pstack/agents/`, attribution, excluded `benny` / `make-bot-ui` | Leave as-is (dual-load / scanner contract) | "Clean up", re-add, or modernize |
@@ -183,23 +183,23 @@ Cursor still appears in three **legitimate** roles only — preserve, do not
 ## Hard rules (CI enforces)
 
 - **Pin lock-step.** `UPSTREAM_PIN`
-  (`93b00b89ef425a9c1bac0d0b317dfc49c930ac99`) is declared in `ci.yml`,
-  `release.yml`, `publish-plugin.yml`, and `upstream-drift-watch.yml` — all four
-  move together. The build steps that consume it are shared via
+  (`c47b12849e43f18d5c374c7069c744cc55b0ea00`) is declared in `ci.yml`,
+  `release.yml`, and `upstream-drift-watch.yml` — all three move together. The build steps that consume it are shared via
   `.github/actions/build-package`.
   Procedure: [docs/RUNBOOK-upstream-drift.md](docs/RUNBOOK-upstream-drift.md).
 - **Anchors audited.** A converter transform whose anchor matches nothing in
   the upstream build fails loudly. Re-anchor or prune; never silence the audit.
-- **Two version lines.** Package keeps upstream identity `0.14.8`; repo
+- **Two version lines.** Package keeps upstream identity `0.15.5`; repo
   releases are `v0.4.x` in `CHANGELOG.md`. Never mix them.
 - **Encoding.** LF-only, no BOM, UTF-8 (`.gitattributes` + `validate.py`).
 - **Converter changes are documented.** New pass/map → ADAPTATIONS row with
   reviewer check + rebuilt `pstack/` in the same PR.
 - **Excluded upstreams stay excluded.** `automations/benny` and
   `skills/make-bot-ui` omitted (slot filled by repo-owned `hermesbot`).
-- **Dist repo is build output.** `iap/pstack` is written only by CI
-  (`publish-plugin.yml`), append-only. Never hand-push. Never point install
-  docs at this dev repo's root (scanner sees `tools/` ban needles).
+- **Single-source installs.** Installs come from this repo's `pstack/` subdir
+  (`hermes plugins install iap/pstack-hermes/pstack`). The old dist repo
+  `iap/pstack` is retired — never publish there again; never document a
+  root install of this dev repo (scanner sees `tools/` ban needles).
 
 ## Common traps
 
@@ -209,10 +209,9 @@ Cursor still appears in three **legitimate** roles only — preserve, do not
   `tools/assets/model-panel.json` and rebuild; `validate.py` checks they match.
 - "Fixing" Cursor mentions that are attribution or dual-load — correct as-is.
 - Rewriting hermes primitives back into Cursor vocabulary — bans catch this.
-- Blurring install routes: **root** (`hermes plugins install iap/pstack`) keeps
-  `.git` and updates via `hermes plugins update pstack`; **subdir**
-  (`iap/pstack-hermes/pstack`) strips `.git` → reinstall only. Docs must keep
-  them distinct.
+- Install-update mechanics: subdir installs strip `.git` (hermes design), so
+  `hermes plugins update` does NOT work for them — updates are reinstalls
+  (`hermes plugins install iap/pstack-hermes/pstack --force`). Docs must say so.
 - Using Cursor IDE dual-load habits as the implementation or review path for
   hermes delivery — verify with convert → validate → doctor / hermes invoke.
 - Pointing hermes `skill_manage` / learning / self-evolution at **stock**

CHANGELOG.md

@@ -12,6 +12,7 @@ All notable changes to the pstack-hermes port tooling will be documented in this
 - `.github/actions/drift-issue-upsert/`: shared composite action for the idempotent issue-by-title upsert both drift watchers need; the two callers' copies had already drifted apart (different jq expressions for the same intent).
 
 ### Changed
+- **Single-source installs.** The package installs directly from this repository's `pstack/` subdir (`hermes plugins install iap/pstack-hermes/pstack`); the dist-repo publish pipeline (`publish-plugin.yml` → `iap/pstack`) is retired and `UPSTREAM_PIN` lock-step now covers three workflows. Updates are reinstalls (`--force`) - documented in README and the installed `after-install.md`.
 - **Re-pin to upstream `c47b1284` (pstack v0.15.5); `pstack/` rebuilt.** Every converter anchor was re-anchored or pruned against the new upstream text (36 dead anchors at the new pin -> 0; `T8`-`T14` + delegation), **F16 retired** (upstream now ships the slug-agnostic lane check itself), `reflect`'s 4-column reviewer table adopted with the upstream `budget` key in the setup-written panel shape, and the two new upstream principles ported (`principle-attack-the-premise`, `principle-test-behavior-not-implementation`) - 45 -> 47 skills. Prose model defaults refreshed (`grok-4.7-xhigh-fast`, `claude-opus-5-5-max`); the shipped panel keeps its vendor-qualified verified slugs. `UPSTREAM_PIN` bumped in all four workflows (lock-step enforced by `tools/check_pins.py`); details and the triage correction (no `skills/orchestrate` ever existed) in `docs/ADAPTATIONS.md` section 4.
 - Docs / agent policy (Phase A, no package rebuild): `AGENTS.md` and `docs/USING.md` codify the stock-skills write surface — Hermes self-improvement (`skill_manage` / learning) targets local skills, skill taps, or a dist fork, not stock plugin skills. Stack forge freeze: **Graphite (`gt`) default**, **GitHub (`gh`) fallback**, **GitLab (`glab`) halted**. README Known limitations aligned (was incorrectly describing Graphite as optional-only).