flow-implement
Skill0 starsChanged 3 months agoApache-2.0View on GitHub
Flow phase 5 (implementation). Normally invoked via /flow. Executes plan.md task-by-task with per-task commits and live progress/deviation tracking in workflow.md.
In this repository9 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- flow-planSkill
- flow-reportSkill
- flow-researchSkill
- flow-reviewSkill
- flow-shapeSkill
- flowSkill
- flow-startSkill
- flow-wrapSkill
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 1 is distinctive, and 3 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 8
Before the first task: copy plan.md's task list into workflow.md ## Progress as unchecked boxes (one per plan task), if not already there — …
- Distinctive: line 16
Run the task's verification step from plan.md.
- Very specific: line 25
Delegation: if .flow.md maps the task's file zone to a domain agent, dispatch that agent with: the plan task text verbatim, the worktree …
- Very specific: line 31
When every Progress box is ticked: run the project verify command, set phase: code-review, update Next action, announce, invoke the …
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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
Flow — implement
Before the first task: copy plan.md's task list into workflow.md ## Progress
as unchecked boxes (one per plan task), if not already there — replace the
template's placeholder comment under the EXISTING heading; never add a second
## Progress.
Per plan task, in order:
- Implement exactly what the task says. TDD where the task has testable logic: failing test → minimal code → green.
- Run the task's verification step from plan.md.
- Commit (
<type>: <plan task title>). - Tick the task in
## Progress(add the commit sha) and bumpupdated.
Deviations: the moment reality diverges from plan.md (extra file, changed
approach, dropped step), record one line in ## Deviations — in the same
turn, not later. Big divergence (approach invalidated, scope grew) → stop,
set phase back to planning, tell the user why.
Delegation: if .flow.md maps the task's file zone to a domain agent,
dispatch that agent with: the plan task text verbatim, the worktree path +
branch, and "report files touched + verification output". Verify its report
against git diff before ticking. Otherwise implement inline — one coherent
context beats a fleet.
When every Progress box is ticked: run the project verify command, set
phase: code-review, update Next action, announce, invoke the
flow-review skill.
Resume: on entry, trust ## Progress — start at the first unticked task.
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.

