The governed coding harness
then run srooterctl login →
Govern every agent change before it reaches production.
One harness for every agent and model. You set the policy. Every request goes through Palgu, which governs model routing, budgets, policy, review and audit for your team. Use it with srooterctl in the terminal or Palgu Studio on the desktop. Your code context and review standards apply to each task.
$bash
>PowerShell
▸add partial refunds to the billing API
HARNESS
contextmapped blast radius → payments · ledger · invoicesCortex code graph
riskHIGH · money + auth path
practicetests required before implementationyour org standard
modelglm-5.3 implements · claude-fable-5.1 reviewsby risk tier
validatebuild ✓ e2e ✓ visual ✓Validation Center
reviewcouncil · 5 seats · 1 blocker caught and fixedindependent models
1 blocker caught: refund > captured amount → fixed before review
One request. Your context, standards and policy, applied automatically.
CODE WITH
srooterctlStudioor route Claude Code, Codex, Cursor, aider
RUNS ON
58×6 frontier models met the same acceptance criteria. Costs varied 58×. See the benchmark →
AnthropicOpenAIGoogleDeepSeekGLMKimilocal · Ollama
THE PROBLEM
Coding agents scaled faster than the controls around them.
Each developer sets up coding agents differently. Platform teams need shared model policy, budgets and a record of what ran.
Palgu applies your team's controls through one API.
Every developer assembles their own AI setup→One harness, defined by your platform team
Each agent sees different context→The right code context per task, from a live code graph
Architecture knowledge scattered in docs→ADRs and codemaps delivered to the agent
Session history is hard to inspect later→A recorded per-session memory trace you can inspect
Models chosen ad hoc, by whoever is driving→Model chosen by task, risk and policy
Agent activity is hard to inspect→Audit records and review results for each task
srooterctlStudioClaude CodeCodexCursoraideryour agents
Palgu harnessowned by your platform team · applied per task
KnowledgeCode graph, blast radius, ADRs, codemaps
MemoryPer-session trace: goal, edited files, errors
PracticeTDD, architecture and review standards
ValidationFunctional and visual checks on a change
ReviewIndependent multi-model council
Model policyRouting, budgets, allowlists, audit
AnthropicOpenAIGoogleDeepSeekGLMKimiOllama · localyour models
Developers pick the client. Your team sets the rules.
HOW IT WORKS
Install srooterctl and start coding.
01 · Install
One install, one sign-in.
Run srooterctl for a coding session, or srooterctl -p to run one task and exit. Every request goes through the Palgu API.
# sign in once, then code
srooterctl login
srooterctl -p "fix the failing test"
srooterctl login
srooterctl -p "fix the failing test"
02 · Apply your standards
Your context, practice and model policy, per task.
Palgu assembles the right code context, enforces your TDD, architecture and review practice, and picks the model for the task.
policy: tests-before-code
context: code-graph + ADRs
model: by task and risk
context: code-graph + ADRs
model: by task and risk
03 · Ship with evidence
Validate and review before the PR.
Functional and visual validation, a review by models that did not write the code, and an audit record for each change.
✓ validated · reviewed
✓ 8/8 checks
✓ audit 7f3a2b recorded
✓ 8/8 checks
✓ audit 7f3a2b recorded
THE WALKTHROUGH
Example: one task, end to end.
A developer types add partial refunds. Here is everything Palgu does before the PR opens.
The model writes the code.
Palgu runs the other eight steps.
- 01Maps the blast radiusResolves the change fan-out: payments, ledger and invoices. Selects the tests that matter.Cortex code graph
- 02Assembles the architecturePulls the relevant ADRs and codemap, including the append-only ledger invariant.Codemap · ADR-041
- 03Grades the riskMoney movement on an auth path. Tier: high. An independent risk model scores it, blind to the solver.risk pipeline
- 04Requires tests firstYour TDD standard applies: failing tests before implementation.org policy · tests-before-code
- 05Picks the model for the tierImplementation on the efficient default. High-risk tasks get a frontier model as reviewer.blast-radius routing
- 06Implements against the testsThe agent works inside the assembled context while Palgu records a session-memory trace you can inspect.Mnemos
- 07Validates functionally and visuallyBuild, end-to-end and screenshot proof attached to the change.Validation Center
- 08Convenes independent reviewFive seats (architect, backend, frontend, tests, security) on models that did not write the code. One blocker caught and fixed.Council
- 09Records the outcomeContext, model, tests, review and outcome in the audit ledger. PR #482 opens with 8/8 checks.audit + provenance
ONE HARNESS · THREE OUTCOMES
Your engineering standards, applied to agent-written code.
CODE GRAPH · billing-svc2,418 symbols · 11 ADRs
impactpayments/refund.py → ledger/post.py → invoices/render.py
adrADR-041 idempotent money movements
dangerledger.post is append-only
teststests/payments/test_refund.py (required first)
reviewcouncil: architect · backend · tests · security · red-team
verdictAPPROVE after 1 fix
HOWlive code graph · org skills · multi-model council · functional and visual validation
ARCHITECTURE AND DEPLOYMENT
Centrally managed. Runs where you need it.
Palgu servers never run your code: tests run in your CI or on the developer machine. The audit log stores a SHA-256 hash of each prompt, not its text. Conversation transcripts are stored separately, with retention your admins control (30 days by default), and encrypted on the managed service (self-hosted: depends on your key setting).
SaaSSelf-hosted gatewayBYOKLocal / sovereign modelsSSO (OIDC · SAML)Audit exportData residencyFull VPC harness · roadmap
Read the platform-team checklist →
agents
srooterctlStudioClaude CodeCodexCursor
runs ondev machines · your CI · your VPC
harness
PalguSaaS or self-hosted gateway
policyallowlists · budgets · SSO · audit ledger · residency
models
frontieropensovereign / localBYOK
WHERE IT RUNS
In your terminal, in CI and on your Mac.
srooterctl runs on macOS and Linux, on developer machines and CI runners. Studio puts Tasks, the Validation Center, the Council Chamber, Knowledge and Routing in one macOS app.
srooterctlThe harness in your terminal, for macOS and Linux. Use it instead of claude or codex: every request goes through your gateway.Install →
Studio for macOSTasks, Validation Center, Council Chamber, Knowledge and Routing in one desktop app.Download for Mac →
CI and scriptssrooterctl -p runs one task and exits. srooterctl review exits 1 when the council requests changes.Download →
Your existing toolsKeep Claude Code or Codex: srooterctl enable routes them through Palgu (opt-in). Cursor and aider connect to the endpoint.See integrations →
PROOF
Measured cost and quality, per model.
We build a real app through the harness with pinned models and score it on four axes: accuracy, speed, cost and quality. The results drive our routing defaults, and they are public.
58×Cost spread, same acceptance criteriaSix frontier models built the same app to the same acceptance criteria.
4Axes scoredAccuracy (Playwright), speed, cost (tokens by price), quality (screenshots).
12Providers, incl. localAnthropic, OpenAI, Google, DeepSeek, GLM, Kimi, Groq, Ollama and more.
5Independent review seatsArchitect, backend, frontend, tests, security, plus an optional red team.
Benchmark conditions, task, commit, model versions and per-run costs are published with every result.See the live benchmarks →