MCP Servers · kernellord/pickysteve

pickysteve

Skill router and context picker for coding agents — picks the right skill per prompt, scans both the request and every retrieved doc for prompt injection.

CLEARED_STATICmcp-portablev0.1.0

Pipeline results

Coverage

CLEARED_STATIC

Tests run

7 / 8

Portability

MCP-portable

Type

MCP Servers

Tests & results

What the pipeline checked

TestStateResult or reason

eligibilityranall eligibility checks passed

dedupranunique -- no near-duplicate found

clocran8584 code lines / 159 files

secretsran0 PLAUSIBLY-LIVE / 2 FLAG-INFO

egressran10 lexical match(es) — FLAG-INFO (C-01: semantic analysis deferred to v2)

injectionranno patterns matched (confidence: STANDARD)

required-fieldsranrequired fields present: name, version, description

behaviouraldeferredbehavioural assessment deferred (v1): Python with non-stdlib imports (committed-only policy): [('tests/test_fixes.py', 'stackone_defender'), ('tests/test_fixes.py', '__future__')]

What's inside

Components found

registry/accessibility-audit.md

componentmedium

registry/api-design-reviewer.md

componentmedium

registry/brand-voice.md

componentmedium

registry/cfo-advisor.md

componentmedium

registry/competitive-landscape.md

componentmedium

registry/create-pitch-deck.md

componentmedium

registry/database-migrations.md

componentmedium

registry/deep-research.md

componentmedium

registry/defi-amm-security.md

componentmedium

registry/docker-optimizer.md

componentmedium

registry/incident-response.md

componentmedium

registry/marketing-campaign.md

componentmedium

registry/performance-profiler.md

componentmedium

registry/postgres-optimizer.md

componentmedium

registry/prompt-injection-defense.md

componentmedium

registry/react-reviewer.md

componentmedium

registry/rust-build-resolver.md

componentmedium

registry/rust-reviewer.md

componentmedium

registry/security-reviewer.md

componentmedium

registry/seo-audit.md

componentmedium

Coverage

Not assessed

behavioural execution not run
egress patterns logged (FLAG-INFO): 10 lexical match(es)
secret pattern(s) classified FLAG-INFO: 2 match(es)

README

From the repository

  



  Picky about what he loads into context, including what he refuses to load.


### ▶ Watch the trailer

https://github.com/user-attachments/assets/8750946b-36be-4c48-bf73-79513451d1f5


  
  
  


PickySteve is a lightweight orchestration layer. A cheap model figures out which skill a request
actually needs, retrieves that one skill, and hands a small, focused, untrusted-data-boundaried
context bundle to a capable model. It does not dump every tool and document you own into context
on every request.

This repo is Phase 1 (MVP), built to an architecture spec. Phase 2 work (tracing platform,
standing eval harness, credential vault, sandbox) is not built yet. Each piece gets added only
when a real Phase 1 failure justifies it.

## 30-second quickstart

```bash
# from the repo root (uv 0.10+; on Windows the venv python is .venv/Scripts/python.exe — substitute it throughout)
uv venv --python 3.11 .venv
uv pip install --python .venv/bin/python -r requirements.txt

# choose your model — local Ollama, OpenAI, Claude, OpenRouter, or any OpenAI-compatible endpoint
.venv/bin/python -m pickysteve.setup

# calibrate the reranker floor on the labeled set
.venv/bin/python eval/calibrate.py…

← Back to MCP Servers