changelog
Skill331 starsChanged 4 months agoApache-2.0View on GitHub
Generate a Discord-ready "What's New" changelog for phase.rs from recent git history. Use when Codex is asked for a changelog, release notes, Discord update, recent shipped changes, or the next sequential changelog batch from a prior tip hash.
- Reads credentials
165 other projects use a version of this.
Adopted by
Most recently taken up by
- deg/beads-utilstheir own version, first seen Nov 16, 2026
- raphaeltm/simple-agent-managertheir own version, first seen Nov 15, 2026
- lengrufeng2008/dsh-skill-librarytheir own version, first seen Nov 15, 2026
- Kuan-AI-com/agent-skillstheir own version, first seen Nov 15, 2026
- chuanglibs/Claude-Code-Game-Studios-Codextheir own version, first seen Nov 15, 2026
- rust-works/omni-devtheir own version, first seen Nov 15, 2026
Used with
Projects with it also use playtest-report, soak-test and 10 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
CoralGameSkillplaytest-report
Generates a structured playtest report template or analyzes existing playtest notes into a structured format. Use this to standardize playtest feedback collection and analysis.
Also in 13 projects that use this
lilangMaxSkillsoak-test
Generate a soak test protocol for extended play sessions. Defines what to observe, measure, and log during long play sessions to surface slow leaks, fatigue effects, and edge cases that only appear after sustained play. Primarily used in Polish and Release phases.
Also in 10 projects that use this
lilangMaxSkillteam-live-ops
Orchestrate the live-ops team for post-launch content planning: coordinates live-ops-designer, economy-designer, analytics-engineer, community-manager, writer, and narrative-director to design and plan a season, event, or live content update.
Also in 10 projects that use this
CoralGameSkillteam-audio
Orchestrate audio team: audio-director + sound-designer + technical-artist + gameplay-programmer for full audio pipeline from direction to implementation.
Also in 13 projects that use this
- +6 more from lilangMax
- +2 more from CoralGame
Other versions165 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 165
- meteor/meteor★ 45ksame name
- Donchitos/Claude-Code-Game-Studios★ 39ksame name
- MaaAssistantArknights/MaaAssistantArknights★ 25ksame name
- pipecat-ai/pipecat★ 17ksame name
- elie222/inbox-zero★ 13ksame name
- hashicorp/terraform-provider-aws★ 11ksame name
- JetBrains/ideavim★ 10ksame name
- BuilderIO/agent-native★ 7.7ksame name
- UnigramDev/Unigram★ 5.5ksame name
- entireio/cli★ 5.3ksame name
155 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository33 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- add-ai-feature-policySkill
- add-card-data-pipelineSkill
- add-engine-effectSkill
- add-engine-variantSkill
- add-frontend-componentSkill
- add-interactive-effectSkill
- add-keywordSkill
- add-replacement-effectSkill
- address-reviewSkill
- add-static-abilitySkill
- add-triggerSkill
- ai-duelSkill
- audit-card-parsingSkill
- batch-mechanicsSkill
- bug-coverage-classifierSkill
- bug-triageSkill
- card-testSkill
- casting-stack-conditionsSkill
- engine-implementerSkill
- engine-plannerSkill
- oracle-parserSkill
- parser-velocitySkill
- pr-contribution-handlerSkill
- pr-flow-auditSkill
- project-referenceSkill
- pr-review-loopSkill
- retrain-ai-weightsSkill
- review-engine-implSkill
- review-engine-planSkill
- ship-commitsSkill
- unlock-setSkill
- validate-cr-annotationsSkill
The file
We split this skill into the rules it gives an agent
12 rules. 0 are common, 6 are distinctive, and 6 are very specific to this project.
0
Common
Most of the 154 edited versions we read keep these.
6
Distinctive
This author's own choices. Worth a look.
6
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 12
Always run git fetch origin first and use origin/main unless the user names another branch.
- Very specific: line 14
Do not use git log --since as authoritative in this repo.
- Distinctive: line 16
Commit hash or tag: use graph reachability.
- Distinctive: line 26
Empty input: default to the last seven days using the same epoch-filter method.
- Distinctive: line 28
If no timezone is supplied, assume Mountain Time and state the offset used.
- Distinctive: line 32
Read commit bodies, not only subjects, for every non-obvious commit:
- Distinctive: line 44
Use emoji-headed sections only when they have content, usually in this order:
- Distinctive: line 53
Use player-facing language.
- Very specific: line 60
After producing the Discord block, emit the SAME content as one structured entry appended to the in-app changelog so …
- Very specific: line 114
After the entry and meta are written, mirror the SAME entry out to Discord so the announcement and the in-app modal stay one source of …
- Very specific: line 124
Always run --dry-run first and show the output before posting — a real post to #announcements is outward-facing and pings members.
- Very specific: line 144
Commit changelog.json, changelog-meta.json, and (when the post ran) scripts/changelog/state.json.
“Common” is measured: kept by at least 6 in 10 of the 154 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Changelog
Generate a user-facing Discord changelog from git history. The input may be empty, a date/time, a commit ref, a tag, or a sequential "next batch" request.
Range Selection
Always run git fetch origin first and use origin/main unless the user names another branch.
Do not use git log --since as authoritative in this repo. Squash-merge commit dates are non-monotonic, and --since can silently drop commits. Use one of these methods:
- Commit hash or tag: use graph reachability.
git log --no-merges <ref>..origin/main --format="%h %s" | cat -n - Date/time: convert the user input to a Unix epoch yourself, then filter
%ct.cutoff=$(date -j -f "%Y-%m-%dT%H:%M:%S %z" "2026-05-22T21:05:00 -0700" "+%s") git log origin/main --no-merges --format="%ct %cI %h %s" \ | awk -v c="$cutoff" '$1 >= c' | sort -rn | cat -n - Empty input: default to the last seven days using the same epoch-filter method.
If no timezone is supplied, assume Mountain Time and state the offset used. Convert named timezones yourself before computing the epoch.
Commit Reading
Read commit bodies, not only subjects, for every non-obvious commit:
git show -s --format="%s%n%n%b" <hash>Cross-check the final changelog against the full commit count so no commit is silently dropped.
Writing Rules
- Output the changelog in a single fenced code block.
- Start with
🎴 What's New in phase.rs. - Use emoji-headed sections only when they have content, usually in this order:
✨ New Cards & Mechanics🛠️ Cards That Now Work Right⚔️ Combat & Gameplay🖥️ Interface🌍 Localization- Other sections such as
🤖 AIor🌐 Multiplayerwhen warranted.
- Use
•bullets. - Consolidate related commits; do not mirror commits one-to-one.
- Use player-facing language. Avoid implementation descriptions unless needed for clarity.
- Name concrete cards or mechanics in parentheses when helpful.
- Order by user impact.
- Omit internal-only changes unless they have visible impact.
In-app changelog entry (client/public/changelog.json)
After producing the Discord block, emit the SAME content as one structured entry appended to the in-app changelog so preview/staging/production surface it too. One run → both outputs. One entry per batch.
client/public/changelog.json is the canonical, committed feed (newest-first,
ids append-only and never reordered). Prepend a new entry:
{
"id": <(entries[0].id ?? 0) + 1>,
"date": "<run date, YYYY-MM-DD>",
"title": "<short specific headline — see title rule below>",
"tags": [<one tag per non-empty section, see map below>],
"body": "<the emoji-sectioned bullets, verbatim from the Discord block but WITHOUT the `🎴 What's New in phase.rs` header line>",
"discordUrl": "<optional link to the matching #announcements post>"
}Section emoji → tag (the tags array mirrors which sections the body contains):
| Section | Tag |
✨ New Cards & Mechanics | new-cards |
🛠️ Cards That Now Work Right | card-fixes |
⚔️ Combat & Gameplay | gameplay |
🖥️ Interface | interface |
🌍 Localization | localization |
🤖 AI | ai |
🌐 Multiplayer | multiplayer |
The tags union is closed (the frontend has a tag→label/color lookup over
exactly these values) — do not invent new tags. The body is rendered as plain
text (newlines preserved); no Markdown/HTML.
Title rule: lead with the batch's single most notable item — a new
mechanic, format, or marquee fix — phrased as a specific ~3–8 word headline
(e.g. "Planechase format arrives", "Dark Depths makes Marit Lage",
"Stickers land, plus 30+ mana sources that finally tap"). Do NOT generate a
tag-join like "New cards & Card fixes" or "Gameplay, Interface & AI": the
modal already shows colored tag chips, so a tag-named title is redundant and
unscannable. Each title must be distinct from existing ones.
Then regenerate the tiny pointer the app reads on every load:
bun scripts/gen-changelog-meta.tsThis rewrites client/public/changelog-meta.json ({ latestId }) and asserts
the changelog invariants (newest-first, unique ids). CI fails if the committed
meta drifts from changelog.json, so always run it before committing.
Publish to Discord (#announcements)
After the entry and meta are written, mirror the SAME entry out to Discord so
the announcement and the in-app modal stay one source of truth — the post is
reconstructed from changelog.json (🎴 What's New in phase.rs + the entry
body), never re-authored, so the two can't drift:
bun scripts/post-changelog.ts --dry-run # preview the exact message(s)
bun scripts/post-changelog.ts # post for realAlways run --dry-run first and show the output before posting — a real post to
#announcements is outward-facing and pings members. This step is safe to run
unconditionally: it no-ops (exit 0) when DISCORD_BOT_TOKEN or
ANNOUNCEMENTS_CHANNEL_ID is absent, so a token-less environment simply skips
it. It is idempotent — it records lastPostedId in
scripts/changelog/state.json and skips an already-posted entry, so re-running
the skill never double-posts. Long entries are split across multiple messages
automatically (Discord's 2000-char limit). When DISCORD_GUILD_ID is set it
writes the post's discordUrl back into the entry so the modal links to it.
Config (same contract as fetch-changelog.ts, no hardcoded secrets/ids):
DISCORD_BOT_TOKEN (gate), ANNOUNCEMENTS_CHANNEL_ID (target channel — NOT
DISCORD_CHANNEL_ID, which is the bug-reports channel), optional
DISCORD_GUILD_ID (for the discordUrl write-back), optional
CHANGELOG_HEADER_EMOJI (the post's leading emoji; defaults to 🎴). A custom
guild emoji is stored bracket-free as name:id (e.g. phase:1500234…) —
the script wraps it as <:name:id> for the API, since a bare :name: only
resolves in the Discord client (not over REST) and the </> brackets would
break source .env. A literal Unicode emoji (no colon) is used verbatim.
Commit changelog.json, changelog-meta.json, and (when the post ran)
scripts/changelog/state.json. The preview snapshot updates on the next push to
main; production picks it up at the nightly release.
Footer
Outside the code block:
- List omitted commits and why they were omitted.
- State the new tip hash so the next sequential batch can use
<tip>..origin/main, and record it inscripts/changelog/state.json({ "lastTip": "<hash>" }) — the generation-side watermark (distinct from the user-facinglastSeenChangelogId).
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

