agentleFS
SentryKnown publishergetsentry/sentry-dotnet/.agents/skills/sdk-review/SKILL.md

sdk-review

Skill776 starsChanged 3 days agoMITView on GitHub

Review sentry-dotnet changes for violations of Sentry SDK contracts: host-app safety, data-collection gates, supported versions, public API, and .NET-specific rules.

6 other projects use a version of this, 3 of them edited.

Adopted by

Most recently taken up by

  • Other versions6 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 3

    1. getsentry/sentry-python★ 2.2k67% same
    2. getsentry/sentry-go★ 1.1k26% same
    3. getsentry/sentry-ruby★ 99045% same

    Wrote their own version · 3

    1. DataDog/dd-trace-java★ 739same name
    2. DataDog/dd-trace-dotnet★ 573same name
    3. getsentry/sentry-godot★ 260same name

    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 repositoryOne other file it gives its agents.

The file

We split this skill into the rules it gives an agent

1 rule. 0 are common, 1 is distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 5 edited versions we read keep these.

  • 1

    Distinctive

    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
  1. Distinctive:

    Follow-ups the PR says are handled in a stacked PR or the major branch.

    line 20

“Common” is measured: kept by at least 6 in 10 of the 5 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.

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

Review only the changed code. Report a finding only if you can point to the exact line and show how it breaks a user's app, build or data. Most hunks break no SDK contract, so no findings is the normal result. Generic logic bugs belong to code-review; this skill only covers SDK contracts.

Core SDK rules (same in every SDK repo)

  1. Never break the host app (high). SDK code that can raise, block, or change the user's behavior: exceptions, return values, signatures, job or request flow.
  2. Data collected without its gate (high). Bodies, headers, cookies, user info, gen_ai inputs/outputs or DB results recorded without their data-collection option. Header and cookie names match case-insensitively. Gate through DataCollectionResolver; SendDefaultPii is only its legacy fallback. Automatically collected header, cookie, query or body values must go through the denylist filter in src/Sentry/Internal/DataCollection/SensitiveDataFilter.cs; a new path that sends them raw is a finding. Data the user set explicitly (scope, span, before_send) is never gated or scrubbed: don't flag reading it back from the scope, but do flag new code that gates or rewrites it.
  3. Supported versions (high). Language, runtime, target or dependency APIs newer than the repo's declared minimum, used without a guard. Minimums come from TargetFrameworks in each src/*/*.csproj (lowest are net462 and netstandard2.0).
  4. Public API (medium). Breaking public changes without a deprecation path, or internals exposed as public API.

Do not report (same in every SDK repo)

  • Style, naming, refactors, or anything the repo's linters, type checkers or analyzers catch.
  • Pre-existing code the diff does not change.
  • Behavior that matches the spec, Relay or other SDKs, or a trade-off the PR or its threads explain.
  • Follow-ups the PR says are handled in a stacked PR or the major branch.
  • Tests, samples and examples, unless they can't fail or they teach wrong SDK usage.
  • Edge cases that need misuse or very unlikely timing.

.NET rules

  1. Customer build breaks (high). Changes in buildTransitive/*.targets or .props that can fail or change customer builds on any OS or target.
  2. Targets (medium). Code that fails on net462 / netstandard2.0, Android, iOS or Mac Catalyst. Check the TFM conditionals.
  3. Options or DI wiring (medium). AddSentry / UseSentry ordering where one overwrites the other's options or callbacks, or options that are never registered.
  4. Undisposed IDisposable (medium). Service scopes, streams or file handles that SDK code creates and never disposes.

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.