Post-performance records — qualification

Record status

Record ID

job:rebecca:post-performance-records

Parent lifecycle

candidate

Page lane

qualification

Lane lifecycle

proposed

Owner

Rebecca

Owner role

Social media operating agent

Owner status

PROPOSED

Qualification result

NOT RUN · held for evidence and human review

Working browser surface: Agent jobs — builder overview — Inspect the package and stage a local project. This route is local-only; provider, credential, database, customer, publication, deployment, and production effects remain held.

Qualification status

Result

NOT RUN

Disposition

HELD

Reviewer

Not recorded

Receipt

Not recorded

This page defines proposed acceptance criteria for the qualification lane. It does not execute tests, establish compliance, qualify the subject, or authorize an effect.

Outcome definitions

Verified

All stated evidence and acceptance requirements passed for the exact reviewed revision; this still does not authorize deployment or an external effect.

Failed

One or more deterministic acceptance requirements failed with an inspectable reason.

Needs review

Evidence or interpretation is incomplete, conflicting, or awaits an identified human reviewer.

Held

Authority, required evidence, safe execution conditions, or a principal gate is absent.

The current record remains NOT RUN / HELD. These definitions describe candidate outcomes only.

Proposed deterministic acceptance tests

Route-resolution and distinct-content checks apply only to the admitted canonical routes in this record. Held source-proposed extensions are non-resolving proposals and are excluded.

• Qualification passes only when the exact post content version has an approval record, the account matches the approved account, chronology fields are complete, every claimed metric has a source locator and capture time, and no negative test is triggered.
• A draft without approval fails as unauthorized and is marked rejected or held.
• Approval for one content version does not qualify a revised content version.
• A scheduled calendar entry does not qualify as published without a receipt or equivalent source-bound publication evidence.
• A publication claim with no timestamp, timezone, source locator, or receipt reference fails the source and chronology tests.
• An observation window ending before it starts fails chronology validation.
• An observation captured before publication fails chronology validation unless explicitly labelled as a pre-publication baseline.
• A metric with a missing value, unit, source, or capture time fails the metric completeness test and is held.
• Conflicting source values are preserved as separate observations and marked needs-review; they are not averaged or resolved automatically.
• A customer name, handle, message, image, or testimonial without an explicit authorization record fails the privacy test.
• Invented customer proof, invented engagement, inferred conversion, or unsupported causal language fails the evidence test.
• Use of an unapproved account fails the authority test even if the content itself was approved.
• Any paid-promotion action or direct-message commitment without separate human approval fails the principal-gate test.
• The record must remain private or access-controlled until a human reviewer approves any public-facing summary.
• Accessibility passes only when every form control has a label, keyboard access, visible focus, non-color status cues, readable error text, and an equivalent text representation for attachments or media evidence.
• A failed external provider request, missing credential, unavailable database, or held upload must produce an explicit held state with retry or human-review instructions and no claim of completion.
• The record must preserve canonical ID, origin, source locator, source hash when available, provenance, revision, lifecycle, links, and attachments without overwriting prior observations.
• The record must remain proposal-only and must not claim deployment, provider connection, publication, customer indexing, or production qualification.

Negative and held-state tests

• Confusing a content calendar schedule with publication evidence
• Self-authorizing publication or treating a native surface name as a connected provider feature
• Recording engagement or customer outcomes without source evidence
• Publishing personally identifying customer information or private direct-message content
• Mixing metrics from different accounts, posts, versions, windows, or timezones
• Overwriting contradictory observations instead of preserving revisions
• Treating an unavailable receipt, credential, database, upload, or provider response as success
• Using color-only status indicators or inaccessible attachment evidence
• Allowing a performance summary to imply causation from observational metrics
• Retaining unnecessary customer data or raw private communications

Qualification claim boundary

Potentially allowed only after the stated evidence

• A draft, approval, schedule, observation, or hold exists in a local proposal record when the corresponding record and state are explicitly shown.
• A metric was recorded from a named source at a stated time when the source locator and capture metadata are preserved.
• A publication receipt was supplied or publication remains held, using the exact recorded state.
• A post was approved for a specified account and content version when a matching approval record exists.

Still held

• The post was published unless publication evidence is source-bound and chronology-complete.
• The post generated customers, conversions, revenue, or positive business impact without verified evidence.
• Any customer testimonial, customer identity, or customer proof is authentic, authorized, or publishable.
• An account connection, credential, database, provider, or external publication path is active.
• A metric is live, complete, current, or representative when its source state has not been verified.
• Rebecca may self-authorize publication, paid promotion, account use, or direct-message commitments.
• The job package is deployed, indexed as customer data, or qualified for production.
• Any causal claim linking the post to observed performance.

Sources and evidence

These pointers bind the page to evidence. They do not grant authority or collapse verification into deployment.

• agents/index.html#rebecca
• source_record:job:rebecca:post-performance-records
• Observed native contract: NotebookCollectionKind project and fieldGuide collections
• Observed native contract: NotebookTodoPayload deadlines and scheduled/deadline views
• Observed native contract: Notebook record provenance, source locators, hashes, revisions, lifecycle, links, and attachments
• Observed native contract: PointGuard context surface and specialist-app ownership
• Observed app routes: /app/notebook/agent-logs/
• Observed app routes: /app/notebook/approvals/
• Observed app routes: /app/notebook/messages/
• Observed app routes: /app/notebook/notes/
• Observed app routes: /app/notebook/files/
• Observed app routes: /app/notebook/projects/
• Observed app boundary: provider, credential, database, customer, publication, and production effects are held