superpowers
claudecodexCore skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques
15 skills · hooks
Port libraries, CLI tools and binaries between languages with proof of parity.
Plugin agentporter from marketplace agentporter in WebEferen/AgentPorter. Commands read from the repo's manifest — 19 days ago.
contractCapture what the original program promises, independent of any language, so a port can be tested against it. Inventories the public surface into contract/api.json, reads the source tests as the executable spec, records the leak list of internals that are observable, and builds an oracle that runs th
depsTurn every dependency of a project being ported into an explicit, evidenced decision for the target language (ABSORB into the standard library, MAP to a package, SHIM with an adapter, PORT, VENDOR, DROP or DEFER). Parses manifests and lockfiles, measures the used surface of each dependency, finds ca
implementPort one unit of an approved AgentPorter plan into the target language, following the unit task checklist (read contract, scaffold, implement in target idioms, mark divergences, port the mapped reference tests, wire the fixtures, get format, lint, typecheck and tests green, review, report, commit).
portPort a library, package, CLI tool or compiled artifact from one programming language to another with evidence at every step. Dependencies get decisions checked against real package registries, the original's behavior is recorded as oracle fixtures, the work is cut into units and implemented one at a
recoverRecover the behavior and public surface of a program you only have as a built artifact (native binary, .NET assembly, JAR or APK, minified JavaScript bundle, Electron app, WASM) so it can be ported to another language. Uses REA (Reverse Engineer Anything) for static evidence and runs the artifact as
statusShow where an AgentPorter port stands (phase, gates, units by status, parity levels, blocked units and the single next action) and, on request, write the port's "Parity with upstream" README section and a release checklist. Read-only and safe to run at any time. Use when the user asks for the status
strategyDecide how a port is built: per-unit strategy (contract port, code port, wrap, drop), the idiom translation map between the two languages, the decomposition into units with layers, waves, a tracer bullet and an order, acceptance criteria, budgets, scope cuts, a risk register and the target scaffold.
surveySurvey a project before porting it to another language. Pins the source revision, detects languages, build, test and lint tooling, measures size, maps the internal module graph, classifies each module's runtime coupling (pure, I/O, runtime-welded, native) and flags license and test-coverage risks, t
targetRecommend the target language for a port. Scores candidate languages on dependency coverage, semantic distance from the source, runtime fit for the project kind, AgentPorter support, where the port will be used and what the team knows, then asks the user to choose and records the decision. Use when
verifyProve parity between a port and the original independently of whoever wrote the port. Runs every recorded oracle fixture through the target harness, resolves every mapped reference test, adds adversarial cases recorded from the original, checks DIVERGENCE markers, assigns a parity level per unit (EX
contract-extractorBuilds the behavioral contract for an AgentPorter port. Writes contract/api.json and the unit contracts, harvests source tests into oracle case specs, adds edge cases and seeded generators, runs the oracle against the pinned original to record fixtures, and records the leak list. Never edits the sou
implementerPorts exactly one unit of an approved AgentPorter plan into the target language, from the unit contract, the idiom map and the recorded fixtures, and stops when format, lint, typecheck and tests are green. Use from the implement skill.
recovererCurates recovered behavior for an artifact-mode port. Reads REA evidence and black-box probe results under .porter/recover/evidence/, runs targeted probes, and edits recover/recovery.json so every interface item and behavior is grouped, evidenced and rated. Use from the recover skill.
reviewerReviews one ported unit's diff against its contract slice and the target language's idioms, and reports findings with file and line. Does not edit. Use from the implement skill.
scoutFinds and assesses candidate packages in the target language's registry for one source dependency of an AgentPorter port. Never names a package that a registry lookup did not confirm. Use from the deps skill.
surveyorRead-only exploration of a source project for the AgentPorter survey phase. Writes .porter/survey.md from survey.json and the code. Never edits the source. Use from the survey skill.
verifierDifferential and adversarial verification of a port against fixtures recorded from the original. Runs the harness, imports reference test results, records new adversarial cases from the original, and computes parity levels with porter.py. Never reads implementer reports or reasoning. Use from the ve
Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques
15 skills · hooks
Rewrite AI-sounding text so it reads like the writer without changing what it says.
1 skill
Up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your coding agent's context.
1 MCP server
Build powerful, stateful CLI interfaces for any GUI application using the cli-anything harness methodology.
5 commands
Automates browser interactions for web testing, form filling, screenshots, and data extraction
12 skills
Use Alibaba Open Code Review from local Codex.
2 skills
This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.