
Security News
Frontier AI Is Now Critical Infrastructure
The Fable shutdown shows how quickly model access can become a business continuity risk for AI-dependent engineering teams.
@agent-pattern-labs/iso-postflight
Advanced tools
Deterministic postflight settlement for AI-agent workflows: reconcile dispatch plans, observed outcomes, artifacts, and post-steps without model calls.
iso-postflight answers: is this dispatched agent workflow settled, and
what is the next safe action?
It is a deterministic, local CLI/library for reconciling a dispatch plan with observed outcomes, artifact evidence, and post-run steps. It does not call a model, does not run an MCP server, and does not add prompt or tool-schema tokens.
Use it after iso-preflight or any other planner has emitted bounded rounds:
npm install @agent-pattern-labs/iso-postflight
iso-postflight status --config postflight.json --plan plan.json --outcomes outcomes.json
iso-postflight check --config postflight.json --plan plan.json --outcomes outcomes.json
iso-postflight explain --config postflight.json
Use --workflow <name> when a config contains multiple workflows, and --json
for machine-readable output. check exits 1 unless the workflow is fully
complete.
iso-postflight status \
--config examples/jobforge-postflight.json \
--plan examples/jobforge-plan.json \
--outcomes examples/jobforge-outcomes-partial.json
Example output:
iso-postflight: STATUS workflow=jobforge.apply
state: ready-for-next-round
next: dispatch-round - Dispatch round 2.
rounds: 1 complete, 1 not-started
outcomes: 2 succeeded, 0 failed, 0 skipped, 0 replacement, 0 in-flight, 0 missing, 0 blocked
round details:
1. complete: job-1=succeeded, job-2=succeeded
2. not-started: job-3=not-started
post:
- merge: pending
- verify: pending
{
"version": 1,
"workflows": [
{
"name": "jobforge.apply",
"successStatuses": ["APPLIED"],
"failureStatuses": ["APPLY FAILED"],
"skipStatuses": ["SKIP", "Discarded"],
"inFlightStatuses": ["running", "in-flight"],
"replacementStatuses": ["APPLY FAILED"],
"requiredArtifacts": [
{
"id": "tracker-tsv",
"label": "Tracker TSV outcome",
"statuses": ["APPLIED", "APPLY FAILED", "SKIP", "Discarded"]
}
],
"postSteps": [
{ "id": "merge", "label": "Merge tracker TSV outcomes", "command": "npx job-forge merge" },
{ "id": "verify", "label": "Verify tracker integrity", "command": "npx job-forge verify" }
]
}
]
}
The plan can be the JSON output from iso-preflight plan --json, or this
minimal shape:
{
"workflow": { "name": "jobforge.apply" },
"rounds": [
{ "index": 1, "candidates": [{ "id": "job-1" }, { "id": "job-2" }] },
{ "index": 2, "candidates": [{ "id": "job-3" }] }
]
}
{
"dispatches": [
{ "candidateId": "job-1", "source": "trace:root" }
],
"outcomes": [
{
"candidateId": "job-1",
"status": "APPLIED",
"source": "batch/tracker-additions/001-acme.tsv",
"artifacts": [
{ "id": "tracker-tsv", "status": "present", "source": "batch/tracker-additions/001-acme.tsv" }
]
}
],
"steps": [
{ "id": "merge", "status": "pass", "source": "npx job-forge merge" }
]
}
import {
formatPostflightResult,
loadPostflightConfig,
settlePostflight,
} from "@agent-pattern-labs/iso-postflight";
const result = settlePostflight(config, plan, outcomes);
console.log(formatPostflightResult(result));
if (!result.ok) process.exit(1);
iso-postflight does not dispatch agents, poll task systems, parse transcripts,
or merge tracker files. Domain code should materialize observations from
authoritative sources such as trace exports, TSV files, ledgers, or workflow
records, then feed those observations into iso-postflight.
Related packages:
iso-preflight checks safety before dispatch.iso-orchestrator owns durable execution and fan-out primitives.iso-ledger stores durable workflow events.iso-trace parses production transcripts.iso-guard audits policy violations after a run.iso-postflight reconciles observed outputs into a next-action gate.FAQs
Deterministic postflight settlement for AI-agent workflows: reconcile dispatch plans, observed outcomes, artifacts, and post-steps without model calls.
We found that @agent-pattern-labs/iso-postflight demonstrated a healthy version release cadence and project activity because the last version was released less than a year ago. It has 1 open source maintainer collaborating on the project.
Did you know?

Socket for GitHub automatically highlights issues in each pull request and monitors the health of all your open source dependencies. Discover the contents of your packages and block harmful activity before you install or update your dependencies.

Security News
The Fable shutdown shows how quickly model access can become a business continuity risk for AI-dependent engineering teams.

Security News
AI agents are pulling packages into environments no scanner is watching, creating exposure before security teams can see it.

Security News
GitHub Actions checkout now blocks risky pull_request_target checkouts by default to help prevent pwn request supply chain attacks.