agentleFS
PostHogKnown publisherPostHog/posthog-foss/.agents/skills/establishing-code-ownership/SKILL.md

establishing-code-ownership

Skill727 starsChanged yesterdayMITView on GitHub

Determine which PostHog team owns a file, directory, or code path, or enumerate all code a team owns (via distributed `owners.yaml`, `products/*/product.yaml`, and `.github/CODEOWNERS`). Use when assigning a reviewer, attributing a bug or slow query to a team, routing work, scoping a team-wide audit, or answering "who owns X" / "what does team Y own".

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 2 are distinctive, and 2 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.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Don't re-parse the ownership files by hand.

    line 9
  2. Very specific:

    No hogli/flox available? The dependency-light fallback needs only pyyaml: git ls-files posthog/hogql | PYTHONPATH=packages/owners-yaml …

    line 24
  3. Very specific:

    If neither the resolver nor product.yaml resolves it, consult the feature-ownership handbook — coarse-grained (broad areas, not files) and …

    line 56
  4. Distinctive:

    If even that fails and the Slack MCP is available, search Slack — least authoritative (opinions, stale threads), so verify against the repo …

    line 57

≈ “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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Establishing code ownership

Ownership is resolved by one tool — hogli owners:* — over distributed owners.yaml files. Don't re-parse the ownership files by hand; the resolver owns the semantics and is what CI enforces.

Fast path: hogli owners:*

Dev machines have flox/hogli, so shell straight to the resolver.

hogli owners:who posthog/hogql/printer.py    # who owns this path (+ the file that decided it)
hogli owners:resolve --json posthog/api/survey.py products/surveys/backend/api.py   # batch, JSON keyed by path
hogli owners:unowned                          # every tracked file with no owner (append a prefix to scope: `owners:unowned products/`)

owners:who prints the resolved owners, the status, the Slack channel, and source — the owners.yaml/product.yaml file that decided the answer. The channel comes from the root teams: registry entry for the primary owner, else a derived #<slug>; a path whose primary owner is an @handle has no channel. owners:resolve takes paths as arguments or newline-delimited on stdin, so you can pipe a file list: git ls-files posthog/hogql | hogli owners:resolve --json. No hogli/flox available? The dependency-light fallback needs only pyyaml: git ls-files posthog/hogql | PYTHONPATH=packages/owners-yaml python -m owners_yaml (stdin paths → the same JSON).

Resolution algorithm (what the resolver does)

For a path, it walks from the repo root down to the path collecting ownership files, then merges them nearest-file-wins:

  1. owners.yaml — the canonical, distributed source. Each directory can carry one. Fields (owners, status, inherit, per-path rules) fall through to the nearest ancestor unless overridden. inherit: false cuts the walk (Gerrit's set noparent) — nothing above it contributes. Within a file, every matching rule applies in order, and each replaces only the fields it sets. owners: null means unowned by design (exempt from the coverage check), distinct from a directory with no file at all (genuinely unowned). additions: names the owners of additions below a directory rather than the owners of the files in it (the consumer decides what counts as an addition); it never changes owners, and it adds up across the walk so a nested file cannot drop an ancestor's entry. sensitive: true marks paths where a small change matters (a CI baseline, an exemption list), so the reviewer assigner requests their owners for every change, whatever its size; it is nearest-wins like status, a product.yaml can set it too, and it never makes a review required.
  2. products/<name>/product.yaml — an accepted alias. The root owners.yaml enables this with alias_files: [product.yaml]. When a product dir has no owners.yaml, its product.yaml owners: list is read as the ownership for products/<name>/**, and its sensitive: flag applies to that tree too (every other product.yaml field is ignored). A dir with both files is a lint error; owners.yaml wins.
  3. .github/CODEOWNERS — blocking approvals, never part of the walk. It keeps GitHub-native semantics, stays hand-maintained (mostly infra, e.g. team-security), and is enforced by GitHub itself. The resolver does not read it — when you need to know whether a blocking approval is additionally required, consult the file directly. It never changes the resolved owners, and nothing here writes to it. A tool that can only read CODEOWNERS gets a generated projection of the map instead, from hogli owners:codeowners (see packages/owners-yaml/README.md); that projection covers test files and is never .github/CODEOWNERS.

Owners are a mixed list of team slugs (team-devex, conversations, logs — the GitHub team handle minus @PostHog/) and @handles for individuals; the first entry is the primary owner.

team → code (what does this team own?)

There's no single "team → files" command; resolve the tree and filter by slug:

git ls-files | hogli owners:resolve --json | jq -r 'to_entries[] | select(.value.owners | index("team-surveys")) | .key'

Also grep products/*/product.yaml for the slug — each hit is all of that products/<name>/**; one team often owns several, so don't stop at the first. Owned paths span backend and frontend/src/...; cover both, or say up front you're doing one side.

Generated files

A generated artifact often resolves to a broad parent owner or nobody. Trace it to the input it's generated from and report that team as the logical owner (e.g. services/mcp/src/tools/generated/<x>.ts comes from products/<name>/mcp/tools.yaml). Distinguish the logical owner (the team owning the source — the answer to report) from the literal resolver result on the generated path, and flag the gap so the operator can decide whether to pin it with a rules: entry.

Last resort: feature-ownership handbook, then Slack

If neither the resolver nor product.yaml resolves it, consult the feature-ownership handbook — coarse-grained (broad areas, not files) and hand-maintained, so prefer the repo files and flag any handbook-sourced answer as possibly stale. If even that fails and the Slack MCP is available, search Slack — least authoritative (opinions, stale threads), so verify against the repo files and flag the answer as Slack-sourced.

Slug vs handle

  • Handle (CODEOWNERS): @PostHog/<slug>, e.g. @PostHog/team-replay.
  • Slug (owners.yaml / product.yaml): handle minus @PostHog/, e.g. team-replay.
  • Not uniform: some carry team- (team-self-driving), some don't (conversations, logs). If a name doesn't resolve, try both forms.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.