Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Get Started Free →

/project-detect - Otomatik Proje Tespiti · vibeeval | DecimalAI

Project Detect

SkillScore22Maintained

project-detect

Projeye girdiginde tech stack'i tespit et, uygun skill ve agent'lari aktive et. Kullanim: /project-detect

agent-toolingby vibeeval

Links this skill to your agents and keeps it current with the author’s updates.

Automatic

Coding pack

1 · One-time — install the CLI (docs ↗)

pip install decimalai

2 · Get the skill — writes .claude/skills/project-detect/SKILL.md

decimalai skills pull project-detect --out .claude/skills/

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/project-detect/SKILL.md https://app.decimal.ai/s/project-detect/SKILL.md

Raw file: /s/project-detect/SKILL.md ↗ · machine-readable summary at /s/project-detect.json ↗

In productionGlobal

how this skill behaves on real traffic · last 30 days

No production data for this skill yet — usage numbers appear here once agents start running it.

Anonymized aggregate across every org using this skill — proof from real production use, not a demo.

Does this skill help? · Benchmark: with vs. withoutsign in to run this on your agents

23 test cases from the publisher’s eval.yaml · each run twice on gemini-3.6-flash — with the skill injected, then without · last run 54d ago

Yes — +22% lift-17% tokens0% turns22% of whole cases pass with skill (5/23)

Test case	Without → With	Effect	Δ tokens	Δ turns
case-10	✗→✓	▲ Improved	—	—
case-11	✗→✓	▲ Improved	—	—
case-09	✗→✓	▲ Improved	—	—
case-03	✗→✓	▲ Improved	—	—
case-12	✗→✓	▲ Improved	—	—

Showing 5 of 23 cases

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 7/30/2026

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

vibeeval / vibecosystem

skills/project-detect/SKILL.md

View on GitHub ↗

License MITImported 8/1/2026Version v1

Skill contents

SKILL.md

/project-detect - Otomatik Proje Tespiti

Mevcut dizindeki dosyalari tarayarak projenin tech stack'ini tespit et.

Adim 1: Dosya Tarama

Su dosyalari kontrol et (paralel):

package.json (name, dependencies, scripts)
tsconfig.json
pyproject.toml / requirements.txt / setup.py
go.mod
Cargo.toml
pom.xml / build.gradle
manage.py
docker-compose.yml / Dockerfile
.github/workflows/
CLAUDE.md (mevcut proje hafizasi)
.env.example

Adim 2: Stack Raporu

PROJE TESPITI
=============
Proje: <dizin adi>
Dil: TypeScript / Python / Go / Java / Rust
Framework: Next.js / Django / Spring Boot / Gin / ...
Database: PostgreSQL / MongoDB / SQLite / ...
ORM: Prisma / SQLAlchemy / GORM / JPA / ...
Test: Jest / pytest / go test / JUnit / ...
CI/CD: GitHub Actions / GitLab CI / ...
Container: Docker / Docker Compose / K8s / ...
Monorepo: Evet/Hayir (turborepo, nx, lerna)

Aktif Skill'ler:
  - <skill-1>
  - <skill-2>
  - ...

Aktif Workflow:
  feature: @architect -> @kraken -> @tdd-guide -> @code-reviewer -> @verifier
  bugfix: @sleuth -> @spark -> @verifier
  refactor: @phoenix -> @kraken -> @verifier

Adim 3: CLAUDE.md Kontrol

Dizinde CLAUDE.md var mi?
Yoksa: "CLAUDE.md olusturayim mi? (template: ~/.claude/templates/CLAUDE-TEMPLATE.md)"
Varsa: Oku, proje bilgilerini kontrol et, eksikleri tamamla

Adim 4: Oneriler

Projeye ozel oneriler sun:

Eksik test varsa belirt
Security riski varsa uyar
Performans iyilestirme firsati varsa soyple
Kullanilmayan dependency varsa belirt

Illustrative Want to see what project-detect would put in front of your model? Preview the rendered body and check its trigger phrases — no LLM call, no signup.

Latest benchmark · publisher runVerified

7/30/2026, 09:56 UTC · 23 cases

Lift

+22%

22% vs 0%

Agent turns

0%

1.0 vs 1.0

Tokens

-17%

1,308.22 vs 1,583.83

fail → pass · 5fail → fail · 18

Per-case results (23)

Case	Status	Duration (ms)			Turns			Tokens			Tool calls
Case	Status	Without	With	Δ	Without	With	Δ	Without	With	Δ	Without	With	Δ
case-04 Write a GitHub Actions workflow YAML file to run pytest and ruff linting on every push to main branch.	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-05 Create a docker-compose.yml file configuring a PostgreSQL database container and a Redis cache service.	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-06 Refactor this Python function to handle database connection errors gracefully using a retry decorator.	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-10 When displaying active workflows in a stack report, what step-by-step sequence of agents is defined for feature development?	fail→pass	—	—	—	—	—	—	—	—	—	—	—	—
case-11 Which agent pipeline should be listed for bugfix operations in the active workflow summary?	fail→pass	—	—	—	—	—	—	—	—	—	—	—	—
case-09 What is the exact agent workflow chain that should be presented for refactoring tasks in a project detection output?	fail→pass	—	—	—	—	—	—	—	—	—	—	—	—
case-03 I just pulled this repo and need a complete technology analysis. Run a directory check on manifest files to identify our language runtime, build systems, test runners, and active skills, then advise on any potential security or test coverage gaps.	fail→pass	—	—	—	—	—	—	—	—	—	—	—	—
case-01 Can you scan the files in the current working directory to analyze what technology stack this project uses? I want a breakdown covering language, framework, database, testing tools, and container setups, along with an overview of recommended development workflows.	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-21 What ORM examples are documented for the ORM field in the project detection report format?	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-08 Which manifest files should be inspected during initial repository scanning to detect JavaScript and Rust projects?	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-15 What specific advice should be surfaced in Step 4 when security issues are identified during repository analysis?	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-16 How should unused packages detected during codebase analysis be communicated in the project report?	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-17 What environment template file in the root directory should be scanned during Step 1 file scanning?	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-12 If during project detection the root directory lacks a CLAUDE.md file, what specific user prompt and template file path should be used?	fail→pass	—	—	—	—	—	—	—	—	—	—	—	—
case-13 When scanning a project that already has a CLAUDE.md file present in the root directory, what actions should be taken?	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-23 Which workflow directory should be scanned during Step 1 to inspect CI/CD configurations?	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-02 Please inspect our root repository configuration files to construct a project detection report. It should summarize our stack details like ORM, CI/CD pipelines, and monorepo status, and let me know if there are any missing setup files or actionable project recommendations.	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-07 Set up a new pyproject.toml configuration using hatchling as the build backend with dependencies for FastAPI and Uvicorn.	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-20 Which module configuration file indicates a Go project during Step 1 file scanning?	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-14 Which monorepo management tools are explicitly evaluated under the Monorepo line of the detection output?	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-18 Which build configuration files should be inspected during initial file scanning to detect Java projects?	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-19 Which setup and configuration files should be inspected to detect a Python project in Step 1?	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-22 What test runner examples are documented for the Test field in the project detection report output?	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—

These cases were also screened by the skill’s intent / safety pass.See the Trust review ↗Reproduce this benchmark in Colab ↗

How this skill was measured

DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 23 cases were attempted. The headline lift of +22 percentage points is the difference between those two pass rates over the 23 comparable cases.

Model measured: gemini-3.6-flash
Graded by: gemini-3.6-flash
Grading method: judged
Verification: verified
Pass rate without the skill: 0%
Pass rate with the skill: 21.7%
Cases attempted: 23
Cases counted toward lift: 23
Cases that regressed: 0
Skill version measured: v1
Run completed: 7/30/2026, 09:56 UTC

The per-case answers from this run were removed by the retention sweep, so the case table below shows the verdicts without the text either arm produced. The counts above were recorded at the time and are unaffected. Answers are now kept for 180 days.

SkillScore blends 4 signals1 of 4 complete

BenchmarkPass rate WITH skill on a verified run22

Live evalNeeds ≥10 evaluated traces in 30 days

AI ratingNeeds ≥10 rated traces

AdoptionNeeds ≥10 serves across ≥2 orgs in 30 days

Backed by 1 of 4 signals — each added signal makes the score more robust.

Verified on gemini-3.6-flash · re-tested 54 days ago

Other measured skills in the registry, with their headline benchmark lift.