Harness E2E

harness-e2e measures what a Harness stack can execute: correct deliverables, structural integrity, bounded work, and repeatable outcomes.

The repository stays independent of the workers source tree. Discovery, execution, observation, state access, and cleanup go through functions registered in iii. The product input is an already-running iii stack.

One binary, harness-e2e, does both jobs. Compose starts it as the asynchronous e2e::* control plane and the injectable Console page. The same binary also runs scenarios and inspects reports from the command line.

Score

A run's score is the sum of the points its evaluated criteria awarded.

• A criterion nobody evaluated adds nothing.
• Scores are not normalized or rescaled.
• Criterion points stay independent of completion and resource limits.
• Criteria do not veto the score or approve a run.
• Completion, technical validity, artifact evidence, and runtime controls are reported separately.
• Infrastructure and execution failures still fail the CLI.

Native criteria keep awards that were already measured when a later check cannot run. That later check has no award and stays not_evaluated. An incomplete criterion set has no total score. A product failure remains technically valid. An infrastructure failure invalidates the run and keeps the observations already recorded.

Every score and every audit flag is deterministic. No scenario calls a second model to judge the first.

Requirements

shell_coder_sandbox, chess_engine_build, and trend_blog prepare their reviewed fixture from an embedded Git bundle. They need Git. They do not need a fixture checkout or HARNESS_E2E_FIXTURE_PATH. Each attempt uses a private workspace, and temporary source checkouts are removed after their contents are read.

typescript_chat_service carries its own frozen skeleton. It needs Node 22.6 or newer on the runner host: the subject's TypeScript application runs through Node type stripping, both in the public suite and in the runner-owned behavioral probe.

The trending-topics build needs Linux amd64, Docker, Git, Python 3, Node, and access to its pinned fixture. See runtime and controls.

The engineering handoff uses a protected disposable checkout of a pinned revision of iii-hq/e2e-fixture.

Develop

From the repository root:

pnpm --dir dashboard install --frozen-lockfile
pnpm --dir dashboard typecheck
pnpm --dir dashboard lint
pnpm --dir dashboard test
pnpm --dir dashboard build
cargo test --locked --all-targets
cargo clippy --locked --all-targets -- -D warnings
node --test tests/dashboard/*.test.cjs
HARNESS_E2E_BIN="$PWD/target/debug/harness-e2e" python3 -m unittest discover -s tests/python -p 'test_*.py'

The Rust build writes dashboard/dist-console/page.js and styles.css, then embeds both in the worker. Node and pnpm must be on PATH.

List scenarios and their definition digests:

cargo run --locked --bin harness-e2e -- list
cargo run --locked --bin harness-e2e -- catalog

Commands

Command	What it does
(none) / worker	Run the Compose-managed control-plane service.
list	Print every scenario id.
catalog	Print the materialized scenario catalog.
run	Execute one or more scenarios against a running stack.
report	Summarize a saved results.json.
models	List models registered in the running stack.
test-plan list	List profile templates and their coverage.
test-plan materialize	Expand one profile into campaigns, groups, and cases.
--manifest	Print the Registry worker manifest as JSON.

Run one scenario against an existing stack:

cargo run --locked --bin harness-e2e -- run \
  --url ws://127.0.0.1:49134 \
  --model codex/gpt-5.6-luna \
  --provider openai-codex \
  --scenario todo_worker_simple

Validate a checked-in campaign without executing it:

python3 scripts/run_e2e_campaign.py config/campaigns/endurance.json --e2e-bin target/debug/harness-e2e --validate-only
python3 scripts/run_e2e_campaign.py config/campaigns/endurance.json --e2e-bin target/debug/harness-e2e --dry-run

Scenarios

Every scenario is a built-in module under src/scenarios/. The module owns its prompt, setup, deterministic evaluator, and cleanup. Its id is the same in the CLI, the worker catalog, the campaign runner, the dashboard, and the canonical result artifacts. Plans name the scenarios they run.

Featured suites:

• Registry: registry_planning, registry_implementation, registry_environment, and registry_verification. Each has its own atomic validations.
• Linkly tutorial: linkly_tutorial runs the seven tutorial chapters plus a project-restart guard as one scripted dialogue on one Harness session, against the linkly-agentic scaffold's Compose stack, and scores twenty-two deterministic checks. scripts/linkly_stack.py prepares that stack.
• Trending topics: an isolated per-attempt Git remote and an independent Playwright acceptance check against the delivered SHA. Design is free. Screenshots are evidence, not an aesthetic score.

config/test-plan.json groups the catalog into modules (state, context, wakes, coordination, software engineering, integration, security, adaptive operations, continuous engineering, and the incremental Kanban application) plus a diagnostic set that profiles can opt into.

Test plans

config/test-plan.json is the executable plan. Rust materializes templates and execution rules from that source and from the native scenario contracts. There is no generated catalog to keep in sync.

Profile	Purpose
smoke	Essential behavior on the published stack.
regression	Representative capabilities, with one technical retry.
capability	Coverage, repeatability, and difficulty across domains.
evolution	Quality and resource use across fixed Harness versions. Three repetitions.
endurance	Sustained correct work and the accepted capability boundary.
software-engineering	Incremental Kanban, Registry, the trending-topics blog, and the Linkly tutorial.

In the Console these profiles are starting templates for the same plan form and the same baseline/candidate view used by saved plans. Choose New plan, optionally pick a template, edit the scope, and select the model. Save draft, Save and run, and Duplicate plan share one lifecycle and keep native evidence. See executable profile plans.

cargo run --locked -- test-plan list
cargo run --locked -- test-plan materialize --profile software-engineering

The software-engineering profile runs each selected case once, with no technical retries: the seven Kanban cases, four Registry cases, the trending-topics build, and the Linkly tutorial. That is 13 cases and 13 planned runs, in 12 execution groups. Registry implementation and verification share a group, in that order, so verification receives the implementation delivery. Trending topics runs in case-trending-topics-build. Linkly runs its eight exchanges in case-linkly-tutorial. The executor creates a fresh pinned linkly-agentic scaffold as that group's Compose project, with baseline worker versions taken from the resolved stack contract.

Release Control

Operational campaigns are dispatched only by Release Control, through .github/workflows/exact-stack-e2e.yml. This repository does not publish independent daily, weekly, post-deploy, or fault-stress dispatch workflows.

Release Control passes five inputs and makes no further decisions: execution_id, the plan (one profile id from config/test-plan.json), a stack policy ({"policy":"latest"} or exact versions), the executor commit runner_sha, and the cli_version to install.

This repository resolves the rest from the commit pinned by runner_sha:

1. harness-e2e test-plan materialize --profile expands the profile into campaigns, groups, and cases, with a profile_sha256 over the result.
2. scripts/resolve_stack_lock.py turns the stack policy into one exact rc-e2e/v2 contract per campaign. Every Registry version is resolved; latest does not survive into a contract. scripts/exact_stack_campaign.py validates that contract.
3. Each group runs in an isolated ephemeral stack. Fault groups go to the protected runner. One root bundle is produced without rebuilding the native Harness artifacts.

Campaign manifests never select or rotate seeds. They keep replay-safe turns separate from scripted dialogue and composite flows, persist a summary for every group, and are advisory while their longitudinal history is being calibrated. Release Control owns scheduling. The executor keeps the result advisory and archives each materialized group through the environment-owned durable archiver.

Partial GitHub reruns reuse the contract artifact produced by prepare, kept for 90 days. The finalizer matches group artifacts to the completed jobs' execution times and keeps successful groups from earlier attempts. A rerun that produces no artifact stays missing evidence; an older artifact does not replace it.

scripts/report_execution.py posts observations to Release Control's run ledger over OIDC: materialized before anything runs, one shard per campaign group, and a summary from the finalizer. Runs come from results.json, or from journal checkpoints when a group died before writing one. A group that produced neither still reports that fact.

workers supplies versioned stack components. It does not orchestrate campaigns.

Agent profile and project template

Release Control can name an existing Directory agent profile in the frozen plan:

{ "agent_profile": "console-ui" }

console-ui is Console UI Engineer. The runner reads that profile from the group's Directory and sends its id as agent to e2e::run. The profile, its parents, skills, functions, and model or provider must exist in that stack. The runner downloads the stack's versioned skill bundles into an isolated Directory and waits up to 120 seconds. A profile that is still missing fails resolution. A profile model overrides the plan model; provider::model also selects its provider. Results record the resolved subject model and the profile configuration hash. Run again resolves the same profile id in its test stack. Comparisons stay manual in Release Control. Omitting agent_profile keeps the built-in agent.

A project template is independent of the test-plan profile and the agent profile:

{ "template": "harness", "agent_profile": "tech-lead" }

The runner reads iii/template.yaml in iii-hq/templates from main, resolves it to one commit before sharding, and records that commit as identity.template.revision. Every group uses that source. The selected Compose project supplies the base. Test-stack versions override its package selectors, extra packages enter the stack lock, and local workers stay local. Template skills replace whole downloaded namespaces, and its agent files take precedence over downloaded profiles. Machine-global profiles and skills are unused when a template or agent override is selected. The runner also enables the campaign's provider when the project does not already have it, honoring any provider version override in the stack contract.

Scenarios, prompts, permissions, fixtures, seeds, and repetitions stay the same. The evaluated agent applies to ordinary sessions and to workflow or adaptive steps. Evaluators stay the same. Linkly keeps its pinned task scaffold and container roles; the selected template's base and agent assets are applied separately. With no template, the run keeps the existing generated stack (or the required fixture).

To measure a profile, compare the same plan, template commit, model, and stack with and without agent_profile. Changing the template as well measures the combined effect. Non-Compose templates and templates that ask for an interactive language choice are rejected before boot.

Console

Build the worker and its page from the repository root:

cargo build --locked --bin harness-e2e

When Console connects to the same iii namespace, the worker registers the page assets and the e2e::dashboard::* functions for read, plan, run, status, and cancellation. The page lives under #/ext/harness-e2e and exposes Overview, Tests, Executions, and Plans. See dashboard/README.md.

The running Harness must publish request and response schemas compatible with the current typed surface. Missing or incompatible fields fail preflight. There is no payload-version compatibility mode.

The page loads through iii: 25 compact summaries on the first overview page, one complete report when an execution is opened, only the selected pair for comparison, and the model and scenario catalog when the run dialog opens. Filtering and cursor pagination run on the server. Transport failures stay visible. The trusted publisher still writes the bounded JSON report archive used by CI. It does not publish a standalone Harness E2E web application.

Compare a local change with Release Control

On Plans, Reference: Release Control browses RC history through the authenticated Release Control browser bridge. Keep the RC tab open, enable its local Harness connection, and connect it to the same personal Engine as the Console. The bridge needs the E2E read functions from the companion Release Control change. No GitHub token or artifact sync is required.

Open a plan to see remote and local executions together, with their origin. Select a reference and a local result to compare measurements. Scenario links open the existing A → B comparison with both executions selected. Missing reports and metrics stay visible as unavailable. Reading history does not create a local plan, and the comparison sends no local results to Release Control.

Run locally on a remote reference saves that execution's materialized test parameters as a local plan and runs them against the current Harness. Repeating the action creates a new local plan from the current scenario contracts and keeps earlier plans and results. Scenarios, rounds, repetitions, and retries come from the execution's materialization, not from a current profile of the same name. The local scenario implementations and Harness are used on purpose: this is a personal experiment, not an exact-stack certification. Fault-injection groups still require the protected executor. References without a shard seed for every scenario cannot be reproduced. Differences in the local scenario definition or case identity are shown as advisory information.

Results stay in the local plan store. The RC execution remains a reference. Native result validation stays strict; remote data is read through the RC API and is not installed as a native report. Full remote evidence is the execution's GitHub link, subject to retention. This flow does not download an evidence archive.

Worker

Release Control names the exact project roots. This repository writes only the root configuration and passes worker@version references to compose::add. iii resolves the Registry graph, writes the project topology, and reconciles containers. Every execution starts an empty Engine and a dedicated Compose daemon, then runs compose::add, compose::up, compose::status, and compose::down. Each execution uses one isolated namespace for Compose and for the project functions it starts.

Compose supplies III_URL, III_NAMESPACE, III_WORKER_NAME, and III_CONFIG. All four are mandatory. The configuration holds the execution-specific evidence directory and the separate control-plane database namespace. Start worker-compose.control.yaml before worker-compose.yaml. The control file provisions the single-connection harness_e2e SQLite pool and disables SQL history. The Harness exits when that database or its schema is unavailable. The subject namespace never receives the database client or its filesystem path.

Publication validates the locally built binary through a path:// Compose container before upload. Published campaigns use only exact Registry package versions. Provider secrets go to temporary permission-restricted env_file files and never appear in contract, Compose, evidence, or archive artifacts.

The worker exposes:

• e2e::run, e2e::status, e2e::cancel
• e2e::results-get, e2e::results-list, e2e::compare
• e2e::scenarios-list
• e2e::archive, e2e::archive-head, e2e::archive-restore
• e2e::history-list, e2e::retention-sweep

Subject policies deny e2e::*.

Live worker check

tests/live_worker.rs talks to a worker registered in a real engine, so those tests are ignored by default. They check that the worker publishes this revision's catalog (ids, seeds, cases, definition digests), that it refuses malformed requests, and, when a subject is named, that one minimal_path execution passes admission, setup, the subject turn, capture, evaluation, and persistence. The resulting report is readable through e2e::results-get:

HARNESS_E2E_LIVE_URL=ws://127.0.0.1:49134 HARNESS_E2E_LIVE_NAMESPACE=my-project \
HARNESS_E2E_LIVE_PROVIDER=deepseek HARNESS_E2E_LIVE_MODEL=deepseek-flash \
cargo test --test live_worker -- --ignored

Seven scenarios state host paths in their prompts, so their definition digests match only when the test process has the worker's HARNESS_E2E_RUN_DIR, TMPDIR, and HARNESS_E2E_*_FIXTURE_PATH values. Export the same environment the Compose file gives the worker. The scenario run leaves one execution labelled live worker validation in the worker's storage.

Storage

Durable artifacts are chunked through storage::*. Admissions, executions, runs, attempts, and artifact references are written through the control-plane database::* worker. Execution records keep compact dashboard summaries and observations, so lists and history do not load native reports.

Storage has no version number and no migration step. Every table records the fingerprint of the statements that create it. At start, the worker recreates tables whose fingerprint moved, in one transaction. It keeps the execution records, local plans, and receipts it can still read, and rebuilds run projections from the native bundles. Rows it cannot read, missing bundles, and imported Release Control history in a recreated table are logged as warnings. That history returns by importing it again. Nothing is reconstructed as a scored result. A report or plan written under another results contract is read with a warning.

Plan definitions and composed execution receipts live in saved_plans and saved_plan_executions. A saved plan or receipt this binary cannot read is deleted on the next read. Plans written by another binary are not migrated.

Release Control history imports use harness-e2e-history, wrapped as {json, sha256} with a sha256: digest of the exact UTF-8 JSON. Import history in the Console imports a file or fetches a plan from the RC bridge. Plans and executions keep source identities, revisions, and every retained report. Repeated imports do not create duplicates. Imported active work never enters local admission or recovery. History stays readable without RC. Evidence uses local gh credentials and Python 3 to verify the GitHub bundle manifest, execution and attempt identity, and file checksums. Missing, expired, inaccessible, and invalid evidence are separate states. Native bundles keep full reports, manifests, and transcripts, loaded on demand. The runner has no S3, GCS, R2, SQL-driver, or Harness dependency.

Weekly Stress materializes deterministic fault plans and evaluates journals from a protected supervisor. Lane promotion is governed by config/policies/cutover.json.

Observation

The runner waits for a session tree to finish by binding harness::turn-completed to an internal sink, e2e::on-turn-completed, before harness::send. That sink is not a control-plane verb: it is not registered with e2e::run, e2e::status, or e2e::cancel, and it does not appear in e2e::scenarios-list. Subject policies already deny e2e::*.

A 15-second watchdog samples harness::metrics and one root harness::status for stuck detection, heartbeat logs, and e2e::cancel. If the trigger type is missing from engine::triggers::list, the run is unsupported infrastructure. There is no fallback that polls harness::status or harness::metrics. After the tree completes, the runner still collects terminal status, metrics, transcripts, and deliverables.

Comparison

Every completed execution records the subject and E2E revisions, observed wire contracts, definition digest, materialized inputs, seed, policies, artifacts, and raw structural evidence. e2e::compare takes two distinct completed execution ids (from_execution_id and to_execution_id) and writes comparisons//e2e-delta.json plus e2e-summary.md. Numeric deltas stay disabled when the case set or the canonical contract differs.

Deliverable, structural, technical, cost, latency, turn, and retry deltas stay independent. A case is repeatable after five local runs meet the deliverable, structural, and technical thresholds. Cost and wall time are observed metrics and are compared only inside a compatible baseline and candidate cohort.

Deterministic assessment has one payload shape, written only to results.json. Scenario contracts are the only versioned domain. Before cleanup, asset capture applies explicit safety limits and writes an unversioned sidecar with the canonical deterministic validation portion, which is aggregated into results.json.

Repository

Path	Owns
src/	Runner, wire adapters, scenarios, evaluation, longitudinal comparison, and the E2E control worker.
config/	Comparison and cutover policies, fault profiles, and the master test plan.
tests/	Test-only fixtures, golden wire schemas, and the Node and Python suites.
schemas/	Public contracts for generated E2E artifacts.
dashboard/	React, TypeScript, Vite, and Tailwind Console page embedded in the worker.

Generated reports, transcripts, logs, and deliverables stay outside Git.

The crate may depend on the iii SDK and on generic libraries. It must not declare a path or Git dependency on workers, Harness, or another product crate. Contract compatibility is established at runtime from engine::functions::list and engine::functions::info. The checked-in schemas are parity fixtures, not a linked product API.

Package boundary

This repository executes exact-stack Test Plans and publishes immutable harness-e2e Registry releases through the manually dispatched release.yml workflow. Release Control supplies a stack policy and an immutable executor SHA to exact-stack-e2e.yml; the contract this repository assembles from them pins every Registry version to an exact one, including historical candidates, because a campaign has to be able to say afterwards what it ran.

The root iii.worker.yaml is the public manifest for local iii worker development and package compatibility. The root worker-compose.yaml is a normal public Compose document. Release Control and post-prepare workflow phases read neither source contract.