dot-ai-changelog-fragment
Skill336 starsChanged 10 months agoMITView on GitHub
Create changelog fragment for release notes. Invoke during /prd-done workflow during the first push to the PR.
In this repository33 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- docs/CLAUDE.mdCLAUDE.md
- dot-ai-generate-cicdSkill
- dot-ai-generate-dockerfileSkill
- dot-ai-impact_analysisSkill
- dot-ai-manageKnowledgeSkill
- dot-ai-manageOrgDataSkill
- dot-ai-operateSkill
- dot-ai-prd-closeSkill
- dot-ai-prd-createSkill
- dot-ai-prd-doneSkill
- dot-ai-prd-fullSkill
- dot-ai-prd-nextSkill
- dot-ai-prds-getSkill
- dot-ai-prd-startSkill
- dot-ai-prd-update-decisionsSkill
- dot-ai-prd-update-progressSkill
- dot-ai-process-feature-requestSkill
- dot-ai-projectSetupSkill
- dot-ai-query-dot-aiSkill
- dot-ai-querySkill
- dot-ai-recommendSkill
- dot-ai-remediateSkill
- dot-ai-request-dot-ai-featureSkill
- dot-aiSkill
- dot-ai-tag-releaseSkill
- dot-ai-usersSkill
- dot-ai-versionSkill
- dot-ai-worktree-prdSkill
- dot-ai-write-docsSkill
- infographic-generatorSkill
- publish-mock-serverSkill
- write-docsSkill
The file
We split this skill into the rules it gives an agent
6 rules. About 0 are common, 5 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈5
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 9
Create a towncrier changelog fragment for release notes when completing PRD work.
- Distinctive: line 15
If not already known from context, ask: "Which PRD should I create release notes for?"
- Distinctive: line 24
Read the entire PRD file to extract:
- Very specific: line 34
Read pyproject.toml to see the available fragment types.
- Distinctive: line 68
If the PRD includes documentation updates, link to the relevant page on devopstoolkit.ai.
- Distinctive: line 101
Link to docs: If PRD updated documentation, link to the specific page on devopstoolkit.ai
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Create Changelog Fragment
Create a towncrier changelog fragment for release notes when completing PRD work. This should be included in the PR so the fragment is reviewed along with the code changes.
Workflow
Step 1: Identify the PRD
If not already known from context, ask: "Which PRD should I create release notes for?"
Look for:
- PRD mentioned in recent conversation
- PRD referenced in current branch name (e.g.,
feature/prd-320-*) - PRD file path provided by user
Step 2: Read the PRD Thoroughly
Read the entire PRD file to extract:
- Problem Statement: What user pain point was solved, why it mattered
- Solution Overview: What the feature does, how it works
- User Impact: Specific benefits, what users can now do
- Key Capabilities: Individual features, options, or modes added
- Technical Details: Configuration options, environment variables, commands
- Documentation Updates: Which docs were added or updated (check Milestones section)
Step 3: Determine Fragment Type
Read pyproject.toml to see the available fragment types. Each [[tool.towncrier.type]] section has:
- A comment above it describing when to use that type
- A
directoryfield (the type identifier used in the filename, e.g.,featurefor.feature.md)
First, ask the breaking question — before matching the change to any other type. Determine whether this change is breaking as this project defines it: does it break compatibility with a prior release, or otherwise require existing users to take action to keep working? Read the project's breaking-type comment in pyproject.toml for the project-specific definition — it may be broader than an API/CLI change (for example a wire-protocol, on-disk-format, or other cross-process/compatibility contract, including a change that stays structurally compatible but shifts meaning). When the change matches that definition — or you are unsure and it plausibly could — prefer the breaking type. Under-classifying a breaking change is costly: it ships with the wrong version bump and with no breaking-change entry to warn users.
Otherwise, choose the type that best matches the PRD based on those descriptions.
Step 4: Write the Fragment
Create file: changelog.d/[issue-id].[type].md
IMPORTANT: Use flat structure, NOT subdirectories!
- ✅ Correct:
changelog.d/329.feature.md - ❌ Wrong:
changelog.d/feature/329.md
Naming convention:
issue-id: GitHub issue number from PRD (e.g.,320)type: Type identifier from step 3 (e.g.,feature,bugfix,misc)
Content format:
## [Feature Title]
[Opening sentence: What this feature is and the problem it solves]
[Key capabilities paragraph: Specific things users can now do, with concrete examples]
[Configuration/usage paragraph if applicable: How to enable or use the feature]
[Documentation link if docs were updated]Documentation links: If the PRD includes documentation updates, link to the relevant page on devopstoolkit.ai. The URL pattern is:
https://devopstoolkit.ai/docs/{project}/{path}- Where
{project}is:mcp(dot-ai),controller(dot-ai-controller),ui(dot-ai-ui), orstack(dot-ai-stack) - And
{path}maps from the docs folder (e.g.,docs/guides/mcp-recommendation-guide.md→guides/mcp-recommendation-guide)
Example: changelog.d/142.feature.md
## Multi-Cluster Management
Manage multiple Kubernetes clusters from a single dot-ai instance. Previously, each cluster required its own dot-ai deployment, making it difficult to compare configurations or apply consistent patterns across environments.
The `query` tool now accepts a `--cluster` flag to target specific clusters, and results indicate which cluster each resource belongs to. The `recommend` tool can generate manifests targeting different clusters with environment-specific customizations. Cross-cluster searches let you find resources across all connected clusters simultaneously—useful for tracking down where a particular workload is deployed. Cluster health aggregation shows a unified view of all clusters in the `version` output.
Configure additional clusters by adding kubeconfig contexts to `ADDITIONAL_KUBECONFIGS` (comma-separated paths). Each context becomes available as a cluster target. The default cluster remains the current kubeconfig context when no `--cluster` flag is specified.
See the [Multi-Cluster Setup Guide](https://devopstoolkit.ai/docs/mcp/setup/multi-cluster-setup) for configuration details and examples.Step 5: Confirm Creation
Show the user:
- The fragment file path created
- The content written
- Reminder to commit and push with the PR
Guidelines
- User-focused: Describe what users gain, not implementation details
- Specific: Include concrete examples of what each capability does
- Complete: Cover all major features added, not just the headline
- Present tense: "Tools now return..." not "Added support for..."
- No diary style: "Multi-Cluster Management" not "Added multi-cluster support"
- Include configuration: Mention environment variables, commands, or setup steps
- Link to docs: If PRD updated documentation, link to the specific page on devopstoolkit.ai
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.

