xp-agile-delivery
Skill23 starsChanged 8 days agoApache-2.0View on GitHub
Use when a story moves from ready to in-progress, or a mission needs slicing. Larger missions: blocked sub-issues in the tracker. Not test layers -> layered-testing-executor. Trade-off doubt -> pragmatic-engineering.
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 49
Read XP choices for a concrete trade-off between testing, pairing, integration and cadence.
- Distinctive: line 51
Use the optional slice note only when the existing work item lacks a useful format.
≈ “Common” is an estimate: at least 1 in 5 of the 308,232 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
XP delivery with useful feedback
Start with a usable result
Read the request, current acceptance and work authority. Express the next slice as a consumer-visible result, not a list of layers. The consumer may be a person, API client, worker or library caller. An API-only delivery is legitimate when that API is the agreed product; a UI feature is not complete merely because its API exists.
Resolve nonblocking uncertainty during the task instead of requiring another discovery document. Trade-off or over-engineering doubt -> pragmatic-engineering.
Slice into the tracker item, never into a plan file
When the mission is larger than one slice, cut it before writing code: name the real entry point and observable result of each slice, and keep a learning spike, an implemented slice and the full requested outcome as three distinct states. The map is the repo's one declared tracker item, with native blocked sub-issues when the tracker supports them (operating rules §5); a local plan file or PRD exists only when the user explicitly asks for one, and then it is the deliverable. Update it when learning changes the next decision, not after every keystroke. A clear low-risk edit needs a short execution sequence, not a plan at all.
Build, test and integrate
Drive new behavior through tdd. Prefer extending an existing meaningful test over creating an overlapping suite. Which layers to run belongs to layered-testing-executor.
Collaborate without manufacturing handoffs
Use a specialist when risk, unfamiliarity or independent work justifies the cost. When pairing, name a driver who writes and a navigator who challenges the diff; hand over write ownership explicitly when switching. A reviewer does not take quality responsibility away from the implementer. Do not force a pair, rotation ceremony or approval chain for every task.
Urgency reduces batch size and optional work, not required acceptance. A deployment-ready change and a deployed feature are different states.
Capture real bottlenecks in the existing tracker; add no velocity dashboard, sprint ledger or retrospective ritual by default.
Read XP choices for a concrete trade-off between testing, pairing, integration and cadence. Read the XP & Agile guide for practice tables and story-splitting patterns when a concrete example is needed beyond this skill's principles.
Use the optional slice note only when the existing work item lacks a useful format.
In this repository21 other files it gives its agents.
- claude-code-orchestratorSkill
- codebase-designSkill
- code-reviewSkill
- codex-subagent-routingSkill
- diagnosing-bugsSkill
- docker-composeSkill
- exhaustive-repo-cleanup-auditSkill
- gemini-cli-orchestratorSkill
- grok-build-orchestratorSkill
- kill-slopSkill
- kimi-cli-orchestratorSkill
- layered-testing-executorSkill
- legacy-code-changeSkill
- pragmatic-engineeringSkill
- refactoring-catalogSkill
- security-reviewSkill
- tddSkill
- test-auditSkill
- testing-e2e-playwrightSkill
- windows-shell-interopSkill
- worktree-lifecycleSkill
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.

