Taskless vs Semgrep
Semgrep is a great engine: a pattern language for rules and a large registry of security rules. The difference is where the rules come from. Semgrep gives you a language for rules. Taskless reads your own review history to decide which rules to write, writes them with tests, and puts each one in the engine that can decide it. Many teams will want both.
Open source CLI. Runs locally, no account required.
Where the rules come from
With Semgrep, rules come from its registry or you write them in Semgrep’s YAML pattern syntax. The registry is the strength: years of security rules you did not have to write. What Semgrep cannot tell you is which rules your own codebase needs.
Taskless starts there. It reads your pull request review comments, fixup commits, and agent files, and finds the corrections your team keeps making. Most teams never write those rules because nobody has time to dig through thousands of PRs. Taskless writes each one with passing and failing fixtures that prove it catches what you meant.
Where each rule goes
When your linter can express a rule, it goes into your linter, in ESLint, Ruff, RuboCop, or Stylelint’s own dialect. Taskless writes its own rule only when your linter can’t: ast-grep for one file’s syntax tree, Vale for the words in docs and comments, and runtime rules (in Taskless Cloud) for checks that read across files. Semgrep matches code with its own engine.
At a glance
| Taskless | Semgrep | |
|---|---|---|
| Built for | Your codebase’s own conventions, enforced on coding agents | Security and bug scanning across codebases |
| Where rules come from | Found in your review history and written for you | You, in Semgrep’s pattern syntax, or the Semgrep Registry |
| Engines | Your existing linter first, then ast-grep, Vale, runtime checks | Semgrep’s own engine |
| Prose and docs | Yes, through Vale | Code-focused |
| Rule tests | Every rule ships with passing and failing fixtures | Supported through annotated test files |
| Getting started | Open source CLI, no account | Open source Community Edition; commercial platform for teams |
Looking at ast-grep or Opengrep as well?
ast-grep is the engine under Taskless’s syntax rules, and the rules Taskless writes are standard ast-grep YAML you can run with the ast-grep CLI without Taskless. Choosing Taskless does not mean leaving ast-grep. Opengrep is an open source fork of the Semgrep engine and sits in the same place as Semgrep in this comparison.
When to use which
Use Semgrep to scan for vulnerabilities and insecure patterns with rules maintained by security specialists. Use Taskless for the conventions your team has decided: which client to call, how errors are handled, what a migration must include, how the docs are worded. taskless check does not run Semgrep, so the two sit side by side in CI without overlapping.
Claims about Semgrep last checked .
Frequently Asked Questions
I already have ESLint, Ruff, or Semgrep. What does Taskless add?
Is Taskless a Semgrep alternative?
Can I use Taskless and Semgrep together?
How does Taskless relate to ast-grep?
Write the rule once. Check it every run.
Install the CLI, then ask your agent to onboard the project. It reads your codebase and history and writes the rules it can enforce locally.