PPromptReports
Use cases/Launch with proof

Launch with proof

Partly builtLDG-2269ContextFileRepoOpsPromptComputeAI Source of TruthPromptReports

Three of five parts merged. The RepoOps release trace and the PromptCompute approval are still to build.

Use case

Launch with proof: a workflow across ContextFile, RepoOps, PromptCompute, AI Source of Truth, PromptReports that starts from a release tag in GitHub, or a person or agent starts it.

Who would use it

Founders and launch teams shipping a release that agents helped build.

What value it provides

Each claim in the announcement links to the change that shipped it and to a dated public record. The docs are checked against the release before anyone writes a word.

Why it matters to the user

Buyers, press and investors repeat launch claims. A wrong one is expensive to chase down; a claim with a record behind it is one less question in every sales call.

The problem

A launch post says what shipped, but nobody checked that the docs match the release, and a claim made on launch day cannot be checked a month later.

Expected outcomes

  • Share of launch claims with a public record: target 100%
  • Docs pages out of date on launch day: target 0
  • Hours from release tag to approved announcement

Each run records its outcome as met, not met or not measured.

How it runs

1. Check each docs page against the release's context versionContextFileBuilt
2. Trace what shipped from the merged changes and their agent sessionsRepoOpsPlanned
3. Draft the announcement, one claim per sentencePromptReportsPlanned
4. A person approves the announcement before it publishesPromptComputePlanned
5. Record each claim with its source changeAI Source of TruthBuilt
6. Show the record badge on the postPromptReportsSwitching on

Products and connections

ProductRole in this workflowDocs
ContextFileCheck each docs page against the release's context versionReference · ContextFile docs
RepoOpsTrace what shipped from the merged changes and their agent sessionsReference · RepoOps docs
PromptComputeA person approves the announcement before it publishesReference · PromptCompute docs
AI Source of TruthRecord each claim with its source changeReference · AI Source of Truth docs
PromptReportsDraft the announcement, one claim per sentence; Show the record badge on the postReference · PromptReports docs

Outside the family: GitHub, Your blog or CMS, A social account (optional).

What a person approves

ActionDefault risk
cms.post.publishhigh
social.post.publishhigh
aisot.record.publishmedium

Action keys are from the PromptCompute catalog.

How to access it from an agent

An agent will run this workflow through the PromptReports MCP server or API, with an ID agent token. The runner will then call each product for it. Proposed

// MCP: https://mcp.promptreports.ai/mcp (Proposed)
start_run({
  workflow: "launch-with-proof",
  inputs: { "release_tag": "v2.8.0", "docs_to_check": "docs/, the changelog", "where_to_publish": "Blog, LinkedIn" },
  dry_run: false
})
// returns { run_id, status: "running", approvals_url }
get_run({ run_id })   // poll, or subscribe to run events
curl -X POST https://api.promptreports.ai/v1/runs \
  -H "Authorization: Bearer $PROMPTREPORTS_AGENT_TOKEN" \
  -d '{"workflow":"launch-with-proof"}'

Or call the products yourself, step by step:

ProductEndpointTools
ContextFileBuiltto confirm
RepoOpshttps://mcp.repoops.ai/mcpsearch_brain get_pre_edit_context propose_brain_entry get_incident_attribution
PromptComputehttps://mcp.promptcompute.ai/mcprequest_approval get_approval cancel_approval answer_approver
AI Source of Truthhttps://aisourceoftruth.com/v1/family/factsto confirm
PromptReportsProposedthrough the runner

Set it up

  1. Sign in at app.promptreports.ai with PromptReports ID, and pick your org.
  2. Connect ContextFile, RepoOps, PromptCompute, AI Source of Truth on the Connections page. Link GitHub, Your blog or CMS, A social account (optional) through the product that uses it.
  3. Register the agent that will run it on the Agents page, and give it a version manifest.
  4. Set the approval policy for cms.post.publish, social.post.publish, aisot.record.publish. Start with Require approval; critical actions are never auto.
  5. Fill in the inputs (Release tag, Docs to check, Where to publish) and choose the trigger: a release tag in GitHub, or a person or agent starts it.
  6. Start a dry run. It will count no workflow request, and it shows which steps would record missing evidence.
  7. Start the run. Approve each gate in the approvals inbox, then read the signed record when it ends.