@jjanczur/tyran 0.1.43 → 0.1.45

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
@@ -11,7 +11,7 @@
11
11
  {
12
12
  "name": "tyran",
13
13
  "description": "A task conductor for Claude Code: multi-agent orchestration with an enforced evidence contract, repo-specific learning, and update-safe local evolution.",
14
- "version": "0.1.43",
14
+ "version": "0.1.45",
15
15
  "source": "./",
16
16
  "author": {
17
17
  "name": "Jacek Janczura",
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "tyran",
3
- "version": "0.1.43",
3
+ "version": "0.1.45",
4
4
  "description": "A task conductor for Claude Code: multi-agent orchestration with an enforced evidence contract, repo-specific learning, and update-safe local evolution.",
5
5
  "author": {
6
6
  "name": "Jacek Janczura",
package/CHANGELOG.md CHANGED
@@ -1,5 +1,129 @@
1
1
  # Changelog
2
2
 
3
+ ## 0.1.45 — 2026-08-22
4
+
5
+ ### The review reads data access and structure, and the self-review has criteria
6
+
7
+ `code-review` gained two dimensions. **Data access and wasted work**: the
8
+ round trips one request makes and how the count moves with the rows — a query
9
+ inside a loop or an awaited per-item `map` is 1 + N, awaits with no dependency
10
+ between them run together with the fan-out bounded, the same pure computation
11
+ runs once per request, and a cache comes AFTER those, never instead of them,
12
+ with the diff saying what invalidates it and why the TTL is that number. None
13
+ of it shows in the diff alone, so the rule is to follow the new call up to the
14
+ request that triggers it, and the finding is a count — "N+1 queries for N rows,
15
+ expected 1" — which is what lets it pin as a test. **Structure**: a
16
+ thousand-line file is built one reasonable addition at a time, so the question
17
+ is whether THIS addition gives a module a second reason to change; the
18
+ counterweight is `deslop`'s own — a layer or helper introduced for one caller
19
+ is the opposite failure, not the cure — and a giant the story merely touched is
20
+ a `NOTES.md` debt, not an in-story refactor.
21
+
22
+ Reported from the operator's own repos: the implementer and reviewer shipped
23
+ N+1 queries, serial awaits over independent calls, repeated queries that could
24
+ have been one, and caches with no stated reason for their TTL, often enough
25
+ that the same eight-point checklist was being pasted into every handoff by
26
+ hand. The checklist now lives in one place — the skill the reviewer already
27
+ loads — and nowhere else.
28
+
29
+ The implementer's "self-review of your own diff" had no criteria: it named no
30
+ skill and listed no dimension. It is now the `code-review` skill run on
31
+ yourself, loaded rather than recalled, with data access called out as the
32
+ dimension an author misses most, because each call site looked fine where it
33
+ was written. For work that reads or writes data the short plan states the
34
+ round trips one request will cost. `deslop` says in one sentence that a slow
35
+ data path is not its business — it deletes, and the fix for a query in a loop
36
+ adds lines — so "the optimization pass" can no longer be read as the
37
+ performance pass. A test pins the pointer and the N+1 clause, because 0.1.40
38
+ is exactly the kind of edit that drops a cross-reference.
39
+
40
+ Not added: a performance-audit skill. The operator's four-phase audit — recon,
41
+ a false-positive pass, a report split by what is safe to fix without asking,
42
+ iterative implementation — is the shape of an initiative under `/tyran:run`:
43
+ scout, plan, implementer, reviewer, with the policy classes deciding what needs
44
+ a human. A skill nothing asks for by name is a library entry, and every
45
+ description is paid for by every session. Skill descriptions: 3868 → 3947 of
46
+ 5000.
47
+
48
+ ## 0.1.44 — 2026-08-19
49
+
50
+ ### The board follows your OS, and finally has a light mode
51
+
52
+ The board was dark only, by declaration in its own docstring. It now ships
53
+ two renderings of the same warm palette — light is the new default face,
54
+ dark keeps its near-black ground — and follows `prefers-color-scheme` until
55
+ you choose. A three-state switch (**System / Light / Dark**) sits in the
56
+ masthead; the choice is stored per browser and applied in the `<head>`,
57
+ before first paint, so a dark-choosing operator never sees a light flash.
58
+ Dark got its contrast fixed while it was open: `--muted`, which carries most
59
+ of the small mono text on the page, was measured at 5.4:1 and raised to
60
+ 6.9:1, and every text role in both palettes now clears WCAG 4.5:1 — computed,
61
+ not eyeballed (the light `--brass` failed at 4.40 and was darkened until it
62
+ did not). Two tokens that had never existed (`--ink`, `--line`) stopped
63
+ silently resolving to nothing.
64
+
65
+ Readability borrowed from a board the operator reads faster than this one:
66
+ one emoji per tab and section label, green/amber/red freshness dots on agent
67
+ chips and stalled cards, a **waiting-time chip** on every open question that
68
+ turns clay after a day, and the recorded DEFAULT promoted from a dl row to a
69
+ bold sentence — "what happens if I never answer" is what the queue is
70
+ actually scanned by. The four headline tiles became buttons that go where
71
+ their number is explained.
72
+
73
+ ### Answering is a chat now
74
+
75
+ The reply box under each question is a composer: the two canned moves —
76
+ **⭐ Use the recommendation**, **Take the default** — as chips above, the
77
+ field and a round send on one line, and **Enter sends** (Shift+Enter keeps
78
+ the newline). The modifier-required rule is retired by operator decision;
79
+ the empty-input guard survives it, so a stray Enter on a blank box still
80
+ sends nothing. A ticket standing in `waiting-operator` embeds the same
81
+ composer in its own detail panel — answering never requires finding the
82
+ question a second tab over.
83
+
84
+ The ticket detail also grew a **timeline** — spawned, reported, reviewed,
85
+ merged, oldest first, merge sha one click from the clipboard — assembled
86
+ entirely from lifecycle facts the payload already carried. Initiative
87
+ documents open from tickets as well as initiatives now, and they render as
88
+ documents: headings, lists and code blocks through a renderer that builds
89
+ structure with `createElement`/`textContent` only. The rule the old `<pre>`
90
+ enforced survives restated — nothing out of an agent-written document
91
+ becomes a link, an image, raw HTML or anything that runs.
92
+
93
+ ### Every setting explains itself before you touch it
94
+
95
+ Each knob on the Settings tab folds open into **"What does this change?"**:
96
+ what the setting is, what changing it does, how it lands on Tyran — and,
97
+ where loosening is possible, what loosening costs, stated *before* the
98
+ confirm dialog restates it as a question. The prose lives in the settings
99
+ catalogue and travels through `settings.json`, so the terminal and the page
100
+ read the same sentences.
101
+
102
+ ### The shipped policy template starts a step looser
103
+
104
+ Operator-decided: new installs got more refusal than they needed. Two dials
105
+ turn, in the template only — an installed policy is KERNEL and no update
106
+ touches it:
107
+
108
+ - `default: GATED` → `AUTO`. The gate is silent outside `.tyran/`,
109
+ `.claude/` and the hooks directory, so the default never protected your
110
+ source tree; it decided only the unnamed paths under those prefixes, where
111
+ an unknown file was denied to every subagent — the exact incident class
112
+ the legacy lease alias records.
113
+ - `.claude/agents/**`: `GATED` → `AUTO`. GATED denies a subagent
114
+ unconditionally, so the retro loop could only ever propose changes to the
115
+ files it exists to maintain. The trade is stated in the rule's own reason.
116
+
117
+ `boundaries:` still ships strict, the KERNEL floors are as unloosenable as
118
+ ever, and tightening either dial back is one word in
119
+ `.tyran/policies/autonomy.yaml`.
120
+
121
+ ### And the board signs its work
122
+
123
+ The masthead links to [janczura.com](https://janczura.com/en/), LinkedIn and
124
+ the GitHub repo; the footer says what the docs site already said — *From
125
+ Berlin with ♥ by two buddies — Jacek and Piotr.*
126
+
3
127
  ## 0.1.43 — 2026-08-18
4
128
 
5
129
  ### One switch that opens the gate, and four things it cannot reach
package/README.md CHANGED
@@ -325,11 +325,11 @@ branch), `tyran:reviewer` (may fix what it finds, but editing forfeits
325
325
  suite on the cheapest tier, reports raw counts, never fixes), `tyran:retro`. A skill is admitted only when something already asks
326
326
  for the protocol by name, and every description is loaded into every session
327
327
  whether the skill fires or not — so the combined length is capped and CI
328
- enforces the cap (3868 of 5000 characters). What each one assumes, when it
328
+ enforces the cap (3947 of 5000 characters). What each one assumes, when it
329
329
  fires and who invokes it, is in
330
330
  [skills and agents](https://jjanczur.github.io/tyran/skills/); the prompts
331
331
  themselves are in [`skills/`](skills/) and [`agents/`](agents/). Behind all of
332
- it: 1646 unit tests, run with `node --test "tests/**/*.test.mjs"`.
332
+ it: 1660 unit tests, run with `node --test "tests/**/*.test.mjs"`.
333
333
 
334
334
  ## Documentation
335
335
 
@@ -54,6 +54,14 @@ commits and anything written to disk are in English.**
54
54
  optimization pass recorded in the story file (`deslop` — it deletes rather
55
55
  than adds, and it needs a test that ran BEFORE your edit), repo validation,
56
56
  then commits, push and PR on the story branch.
57
+ - **The self-review is the `code-review` skill run on yourself** — load it
58
+ with the `Skill` tool; a checklist recalled from memory is the one you
59
+ already failed to apply while writing. The dimension an author misses
60
+ most is data access: each call site looked fine where you wrote it, and
61
+ the loop that multiplies it is in a file you did not open, so follow your
62
+ new call up to the request. For work that reads or writes data, the
63
+ short plan states the round trips one request will cost — a number
64
+ written down before the code is the cheapest one you will ever fix.
57
65
  - When a test fails for a reason you cannot explain, **stop patching and
58
66
  follow `root-cause`**. Reproduce it, change one variable at a time with
59
67
  the prediction written down first, and name the mechanism. A fix for a
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@jjanczur/tyran",
3
- "version": "0.1.43",
3
+ "version": "0.1.45",
4
4
  "description": "Command-line scripts for the Tyran task conductor (doctor, journal, schema, and repo tooling) for CI and terminals outside Claude Code. Installing this package does NOT install the Claude Code plugin — add that separately with `/plugin marketplace add jjanczur/tyran` inside Claude Code.",
5
5
  "type": "module",
6
6
  "engines": {