Skip to main content

Slygentify

Slygentify helps people and coding agents answer practical questions before working in an unfamiliar local Git repository:

  • What applications, packages, and workspace members are present?
  • Which runtimes, package managers, tools, frameworks, and CI commands are declared?
  • Which component owns the file or path involved in a planned change?
  • Which repository facts are missing, conflicting, unsupported, or unsafe to infer?
  • Does managed coding-agent guidance still match the current repository?

It builds a bounded, evidence-backed operating map from static repository evidence instead of guessing from conventions or executing discovered project commands. First-class ecosystem inspection covers Python and JavaScript/TypeScript repositories, including mixed repositories and workspaces.

Slygentify 1.0.0 is the first stable public release. Install the exact release from PyPI or use a reviewed source checkout.

What Slygentify helps with

When you need to... What Slygentify provides Command and interface
Explore or onboard to an unfamiliar repository A component-organized view of identity, declared working tasks, architecture, automation, concerns, and inspection boundaries slygentify scan PATH for a complete human-readable report, or add --interactive for a full-screen terminal explorer
Investigate a large or mixed repository interactively A keyboard-accessible component tree with search, record and classification filters, evidence-first detail, raw record JSON, and a glossary slygentify scan PATH --interactive
Give an agent or automation a complete repository result Deterministic, versioned JSON containing the retained evidence, findings, diagnostics, and skipped scopes slygentify scan PATH --format json
Prepare an agent for one planned change Bounded JSON context for a logical path, including its owning component and selected orientation, workflow, architecture, automation, or boundary records slygentify map PATH --scope FILE
Establish durable repository guidance for coding agents A reviewable plan for concise root AGENTS.md guidance and deterministic provenance slygentify init PATH --dry-run, followed by slygentify init PATH only after review
Detect stale or unsafe managed guidance A fresh comparison of configuration, repository evidence, generated guidance, and managed provenance slygentify doctor PATH
Investigate incomplete or inconsistent repository metadata Explicit diagnostics for malformed, conflicting, unsupported, skipped, or resource-limited evidence instead of silent guesses slygentify scan PATH or slygentify doctor PATH

Human-readable interfaces are designed for exploration and review. Coding agents and automation can consume canonical versioned JSON from scan, map, and doctor, or use the corresponding Python APIs.

What Slygentify understands

Slygentify statically inspects supported manifests, workspace declarations, lock and configuration files, and CI workflows. Depending on the ecosystem and available evidence, it can surface:

  • component and workspace boundaries and their relationships;
  • declared runtimes, package managers, direct dependencies, and entry points;
  • configured development tools and directly declared frameworks;
  • setup, run, test, lint, format, and build commands declared in project metadata or CI;
  • conflicting declarations, malformed supported files, skipped scopes, and important unknowns.

Every conclusion remains classified as verified, inferred, recommended, or unknown, and its supporting evidence remains inspectable. A declared command is evidence that the command appears in repository configuration; it is not proof that the command works or that the team prefers it.

Slygentify does not analyze application business logic, build a symbol or call graph, decide which source files an issue requires changing, resolve dependencies, evaluate dynamic configuration, or execute and verify discovered commands. Sandboxed command verification, cloud services, and runtime CodeGraph or forge integrations are not implemented.

Compatibility

Compatibility means that Slygentify can inspect the named static repository evidence. It does not imply that Slygentify installs the tool, runs its commands, resolves its dependencies, or connects to its hosted service.

Ecosystems and project formats

Ecosystem or format Inspection level Recognized evidence
Python First-class pyproject.toml, supported setup.cfg, uv and Poetry metadata and locks, requirements and constraints files, Python runtime declarations, supported tools, direct framework declarations, entry points, workspaces, and attributable CI commands
JavaScript and TypeScript First-class package.json, npm, pnpm, Yarn and Corepack declarations, Node.js and npm runtime declarations, workspaces, supported tools, direct framework declarations, scripts and bin entries, TypeScript project references, and attributable CI commands
Rust and Cargo Component boundaries Cargo package and workspace boundaries and safe workspace-member relationships from Cargo.toml
Go Component boundaries Module and workspace boundaries and safe workspace-member relationships from go.mod and go.work
Java and Maven Component boundaries Maven project and module boundaries and safe module relationships from pom.xml
CMake and ESP-IDF Project boundaries Static project(...) and idf_component_register(...) markers in CMakeLists.txt; ecosystem-specific metadata remains unsupported
KiCad Project boundaries Valid .kicad_pro project boundaries and associated .kicad_pcb and .kicad_sch artifact evidence
Other ecosystems Explicit declaration A maintainer can declare a component boundary in root slygentify.toml; unsupported ecosystem-specific metadata remains unknown

See the detailed Python and JavaScript/TypeScript inspection references for exact fields, tools, frameworks, and deliberate exclusions.

CI configuration

CI platform Recognized static evidence Important limits
GitHub Actions Workflow YAML under .github/workflows/, literal run commands, supported static Python and Node.js setup values, working directories, and literal checkout paths used for component attribution Expressions are not executed; expression-only commands and ambiguous checkout scopes remain unknown
Gitea Actions Workflow YAML under .gitea/workflows/ with the same bounded static command, runtime, working-directory, and checkout-path inspection Expressions are not executed; unsupported or ambiguous values remain unknown
GitLab CI/CD Root .gitlab-ci.yml, literal script or run fields, working directories, and bounded in-repository include:local files Dynamic and external includes are not fetched; expressions are not evaluated
Other CI systems No platform-specific inspection Files may contribute only evidence supported by another detector or an explicit component declaration

CI compatibility is inspection-only. Slygentify does not connect to GitHub, Gitea, GitLab, or their runners, and it does not verify that a declared workflow succeeds.

Quickstart

Install the exact stable release with pipx install slygentify==1.0.0 or uv tool install slygentify==1.0.0. The installation guide also covers pip and reviewed source checkouts. Every command accepts a directory inside a local Git repository.

Start with the complete human-readable scan report:

slygentify scan path/to/repository

For a searchable full-screen view intended for people working at a terminal:

slygentify scan path/to/repository --interactive

For a coding agent, script, or other automation, select canonical JSON or narrow the result to the logical path involved in a task:

slygentify scan path/to/repository --format json > scan.json
slygentify map path/to/repository --scope src/example.py

When you want durable coding-agent guidance, review the exact proposed artifacts before allowing a write. After initialization, use doctor following structural, tooling, or workflow changes:

slygentify init path/to/repository --dry-run
slygentify init path/to/repository
slygentify doctor path/to/repository

init is the only implemented command above that writes repository files. It writes AGENTS.md and .slygentify/state.json only after planning and revalidation. Existing unmanaged or human-edited guidance is preserved by default; ordinary init prints a paste-ready section and exits 4 so it can be merged manually. Use --adopt --dry-run to preview management of one visible Slygentify section while preserving surrounding human guidance. --replace is an explicit destructive choice and does not create a backup or merge. See the initialization guide for ownership, recovery, and state-schema behavior.

The Git executable is optional. Without it, tracked-path discovery may be unavailable and an otherwise useful result can be partial; diagnostics explain what was omitted.

See the first-repository tutorial for a runnable walkthrough, the troubleshooting guide for safe recovery, and the CLI guide for complete command behavior. The optional root slygentify.toml is described in the configuration reference.

Effects and safety

Core inspection is local, read-only, bounded, contained within the selected Git root, and network-free. Slygentify does not import the target project, execute commands it discovers, read sensitive paths by default, follow descendant symbolic links, or upload repository content.

Automatic scans, including the fresh scan used by doctor, may invoke one fixed, bounded Git ls-files lookup to retain tracked manifests that checked-out ignore rules hide. --git-executable PATH is a separate, explicit authorization for that exact executable. It is trusted unsandboxed code and may have arbitrary file, process, credential, or network effects; inspect nonstandard or repository-contained executables before selecting one. Doctor does not execute discovered validation commands.

A complete scan completed within the supported inspection boundary; it is not proof that every repository fact is known. A partial scan succeeded but omitted or limited some evidence. Diagnostics and skipped scopes explain the boundary. Diagnostic dispositions distinguish problems, trustworthy limitations, and notices without changing claim classification, completion, or exit behavior. Findings preserve whether claims are verified, inferred, recommended, or unknown.

The detailed contracts are in interaction design, inspection accounting, and the accepted safety ADRs.

Documentation

Start with the first-repository tutorial, or use the documentation index to find a focused guide or reference:

Development

Slygentify supports Python 3.11 through 3.14 and uses uv with a committed lockfile.

uv sync --locked --all-groups
uv run pytest
uv run --locked mkdocs build --strict
uv run pre-commit run --all-files
uv build --no-sources

The pytest suite enforces 100% statement and branch coverage. The complete locked command set, Doorstop traceability workflow, branch policy, effects review, and Definition of Ready/Done are in CONTRIBUTING.md. Repository automation runs quality, documentation, supported-Python test, packaging, and dependency-vulnerability workflows; the dependency audit requires network access.

Contributions use short-lived branches from develop and reviewed pull requests back to develop. main receives reviewed release promotions. Coding agents may prepare work and evidence but do not approve or merge pull requests, publish releases, or close human gates.

Slygentify is licensed under the Apache License 2.0.

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

slygentify-1.0.0.tar.gz (120.7 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

slygentify-1.0.0-py3-none-any.whl (141.2 kB view details)

Uploaded Python 3

File details

Details for the file slygentify-1.0.0.tar.gz.

File metadata

  • Download URL: slygentify-1.0.0.tar.gz
  • Upload date:
  • Size: 120.7 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/7.0.0 CPython/3.13.14

File hashes

Hashes for slygentify-1.0.0.tar.gz
Algorithm Hash digest
SHA256 0b67c924948f3fb659d9b2c5b8dcaaf39a63bfd99e83d40b71438ee303678883
MD5 fa64060c5ef974e8dc7f2f58c1b06972
BLAKE2b-256 2951c3a100748356437803e645fd14ea942bed81ef5d5e7413f2a561852815ba

See more details on using hashes here.

Provenance

The following attestation bundles were made for slygentify-1.0.0.tar.gz:

Publisher: release.yml on slytheros/slygentify

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file slygentify-1.0.0-py3-none-any.whl.

File metadata

  • Download URL: slygentify-1.0.0-py3-none-any.whl
  • Upload date:
  • Size: 141.2 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/7.0.0 CPython/3.13.14

File hashes

Hashes for slygentify-1.0.0-py3-none-any.whl
Algorithm Hash digest
SHA256 6a36963fbdd3259e225a59690132d2972165133ce07a43afaa0cf2931b865cc1
MD5 d9d0a3676e15e748e62ab2fce1ae94b5
BLAKE2b-256 2c5f923dc278036fe4889eed3e9f271e64a5e2a499af010690773fb501c9cb81

See more details on using hashes here.

Provenance

The following attestation bundles were made for slygentify-1.0.0-py3-none-any.whl:

Publisher: release.yml on slytheros/slygentify

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Release history Release notifications | RSS feed

This release

1.0.0 This release

2 files

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page