task-pipeline-skill 1.84.0 → 1.84.1

package/CHANGELOG.md

@@ -1,3 +1,41 @@
+## v1.84.1 — the conformance sentence is computed, and two of its five figures were wrong
+
+**A sentence of five measured figures, and nothing derived any of them.** The README's
+*Held to Anthropic's own Skill authoring guidance* paragraph states `name` length,
+`description` length, `SKILL.md` lines, the reference count and the bundle size. Measured
+2026-09-05:
+
+| stated | measured |
+|---|---|
+| `name` 13/64 | 13 ✓ |
+| `description` inside 1024 | 899 ✓ |
+| `SKILL.md` **334**/500 lines | **262** |
+| all **38** references | 38 ✓ |
+| **436 KB** against 30 MB | **1128 KB** |
+
+**The line figure was wrong when it was written.** At commit `297a0cb`, which added the
+sentence, the file was **339** lines — not 334. It was never a measurement, and a reader
+had no way to tell which of the five to trust.
+
+**The size figure changed SHAPE rather than value, and that is the decision worth
+recording.** It moved on four of the eight commits before this one — 950 → 970 → 980 →
+1128 KB — so an exact figure in prose goes stale by construction, and a check over it
+would fire on nearly every edit. *A check that fires constantly is one somebody turns
+off.* It is a bound now — **far under the 30 MB ceiling** — which stays true across three
+orders of magnitude and still says what the sentence exists to say.
+
+The other four are derived and compared on every run. The parser **asserts it found all
+four**: a sentence reworded past it fails here saying so rather than passing quietly,
+which is the shape that has twice made a healthy guard in this family report itself
+broken. Line count is `splitlines()`, stated in the failure message because `wc -l` over
+a file without a trailing newline gives a different number.
+
+Both directions watched failing, each plant asserting it landed first: a wrong figure
+gives *states lines = 999 and the tree gives 262*; deleting the reference clause gives
+*no longer states references in a shape this check can read*.
+
+Guards: 424 → **424** — the new check is an ordinary step, not a planted-defect guard.
+
 ## v1.84.0 — fifteen filed findings, and the two rules they mostly refine
 
 Every one of the fifteen `ssheleg/sshlg-skills` issues this repository owned, closed in one

package/README.md

@@ -587,9 +587,9 @@ must say so.
 ### Held to Anthropic's own Skill authoring guidance
 
 Audited against the four Agent Skills pages. Most of it already held — `name`
-13/64 chars, `description` inside 1024, `SKILL.md` 334/500 lines, all 38 references
-linked **directly** from `SKILL.md`, 436 KB against a 30 MB ceiling. What did not,
-now does:
+13/64 chars, `description` 899/1024 chars, `SKILL.md` 262/500 lines, all 38 references
+linked **directly** from `SKILL.md`, and the bundle far under the 30 MB ceiling. What
+did not, now does:
 
 - **Every reference over 100 lines carries a `## Contents` list**, and the list is
   *compared against the file's own headings* rather than trusted. The guidance is

package/SKILL-CARD.md

@@ -12,7 +12,7 @@ harmless.
 |---|---|
 | **Purpose** | Runs a substantial task through ten gated delivery stages — intake grill, docs study, brainstorm, spec, plan, subagent build, tests, lint/deploy, post-deploy, docs+registers, acceptance — refusing to advance until each gate passes |
 | **Owner** | ssheleg ([github.com/ssheleg/task-pipeline](https://github.com/ssheleg/task-pipeline)) |
-| **Version** | 1.84.0 |
+| **Version** | 1.84.1 |
 | **Surface** | Claude Code (filesystem skill + plugin) and the vercel `skills` CLI. **Not** uploaded to the Skills API; custom Skills do not sync across surfaces |
 | **Dependencies** | None required. Optional: `context7` (MCP), `figma` (MCP), super-ux, agent-sync, graphify, obsidian-wiki, and **one of two browser channels** — `playwright` (CLI or MCP) or `chrome-devtools` (MCP); either satisfies the browser step and neither is required. Every stage's doctrine ships in-repo; the one conditional requirement is super-ux for the stage-3 UX track on a user-facing task |
 | **Evaluation status** | Suite authored, 5 categories. One recorded run, **self-observed by the author**; **zero blind runs on zero of three models** — the split, and the numbers, live in [`evals/RESULTS.md`](evals/RESULTS.md) and are computed by `evals/run.py` |

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "task-pipeline-skill",
-  "version": "1.84.0",
+  "version": "1.84.1",
   "description": "Full-cycle delivery pipeline for coding agents: a mandatory built-in intake grill, then 10 gated stages (docs, brainstorm+decompose, spec, plan, build, tests, lint/deploy, post-deploy, docs/wiki, acceptance). Every stage's doctrine ships inside the skill — no companion plugin required. This package is the installer CLI.",
   "bin": {
     "task-pipeline": "bin/task-pipeline.js"

package/plugins/task-pipeline/.claude-plugin/plugin.json

@@ -3,7 +3,7 @@
   "name": "task-pipeline",
   "displayName": "Task Pipeline",
   "description": "Runs a substantial task through a mandatory built-in intake grill, then 10 gated stages (docs, brainstorm+decompose, spec, plan, subagent build, tests, lint/deploy, post-deploy, docs/wiki, acceptance). Every stage's doctrine is built into the skill — no companion plugin required — with typed auto/judgment/manual gates, a frozen requirement spine that closes with evidence, a work board and a verification ledger that outlive a run, an exposure line naming what shipped unconfirmed, a progress rail computed from the project's own config, a loop guard whose review ceiling measures rather than stops, and stage-3 tracks for what a product does, how it sounds and how it looks. Two modes need no task: `checkup` (what is unverified) and `setup` (audit existing docs). Retro insights can publish upstream as issues, opt-in and redacted.",
-  "version": "1.84.0",
+  "version": "1.84.1",
   "author": {
     "name": "ssheleg",
     "url": "https://x.com/sshlg93"