Source: https://guthlabs.ai/docs/tools/pointguard/qualification/ tool / candidate # Pointguard — qualification The distinct qualification record for Pointguard. This page is proposal-only. Status: **proposed**Lifecycle: **candidate** ## Record status **Record ID**: tool:pointguard **Parent lifecycle**: candidate **Page lane**: docs_qualification **Lane lifecycle**: proposed **Owner**: Tilly **Owner role**: Pointguard and independent reviewer **Owner status**: Not separately recorded **Qualification result**: NOT RUN · held for evidence and human review **Working browser surface:** [Tools — overview]() — Inspect the tool contract, usage, boundaries, and proposed qualification checks. 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 docs_qualification lane. It does not execute tests, establish compliance, qualify the subject, or authorize an effect. ### Lane-specific instrument aids [regex-testeroptional · Deterministically inspect required record fields, route patterns, and prohibited-claim markers.]()[checksum-verifieroptional · Verify that evidence inventories or exported qualification records have not changed between review stages.]()[binary-text-diff-similarity-ratiooptional · Compare draft and proposed-public versions to detect unreviewed claim changes.]() ## 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. ## Catalog-observed source metadata These fields are reproduced from the admitted candidate record for source-fidelity inspection. Presence does not prove implementation, connection, qualification, authority, or an external effect. **tool_id**: tool:pointguard **slug**: pointguard **title**: Pointguard **owner_agents**: Tilly **owner_roles**: Pointguard and independent reviewer **related_jobs**: Independent review; Instrument qualification; Evidence reconciliation; Grant research and drafts; Website claim review **related_field_guides**: Instrument packets; Qualification records; Grant dossiers; Public-claim routes **principal_gates**: Final publication; Funding commitments; Credential access; Changes to another agent's source **never_rules**: Approve her own evidence; Hide a contradiction; Turn a draft into a public claim **lifecycle**: candidate **source_refs**: agents/index.html#tilly ## Proposed deterministic qualification tests - AT-01 Source fidelity: the record must preserve tool_id pointguard, slug pointguard, title Pointguard, owner agent Tilly, all five related jobs, all four field guides, all four principal gates, both never-rules, and source reference agents/index.html#tilly. - AT-02 Proposal boundary: every page and export must state that implementation, connection, qualification, deployment, and authority are unproven unless separately evidenced. - AT-03 Independent review: a reviewer cannot mark a record final when the reviewer is the evidence owner, subject owner, or otherwise has a declared conflict; the record must show an independent reviewer or remain held. - AT-04 Contradiction visibility: any mutually inconsistent evidence must create a contradiction entry with both references, a description, impact, and unresolved status; omission fails. - AT-05 Draft/public separation: a draft claim cannot enter a public-claim route unless a qualification decision, source references, reviewer identity, and publication-gate result are present. - AT-06 Principal-gate lock: final publication, funding commitments, credential access, and changes to another agent's source each require an explicit independent-review decision; missing decisions produce held status. - AT-07 Accessibility keyboard test: every review, evidence, contradiction, test, and decision control must be reachable and operable by keyboard, have a visible focus indicator, and expose its name, role, value, and validation error to assistive technology. - AT-08 Accessibility non-visual test: status, failure, held, and contradiction states must not rely on color alone; headings, landmarks, labels, error text, and reading order must remain meaningful without styling. - AT-09 Privacy test: collect only review data required by the record schema; reject secrets, credentials, unnecessary personal data, and unrelated source content; redact sensitive values in exports and logs. - AT-10 Abuse test: prevent bulk submission, credential capture, impersonation, evidence deletion, reviewer self-approval, hidden contradictions, and unauthorized edits to another agent's source; rate-limit and audit repeated failures. - AT-11 Source test: every allowed claim must point to an exact supplied source reference or an attached evidence record; absent provenance means held, not inferred. - AT-12 Authority test: the system must never represent Pointguard, Tilly, or an independent reviewer as having implementation, deployment, credential authority, funding authority, publication authority, or source-edit authority unless separately proven. - AT-13 Determinism test: identical normalized inputs, evidence references, and test versions must produce the same status, failure list, and decision rationale. - AT-14 Instrument-directory test: compatible instruments must use only exact slugs from the supplied allowed directory; unlisted or disposed instruments must be rejected. - AT-15 Export integrity test: exported records must include revision identifier, record hash or equivalent change marker, source references, test results, held claims, and audit history. ## Negative and abuse cases - Self-approval or conflicted review - Unsupported conversion of drafts into public claims - Hidden or discarded contradictions - False implication that the observed catalog entry proves implementation or authority - Unauthorized credential access or source modification - Exposure of submitted secrets or personal data - Evidence tampering or untraceable revisions - Accessibility barriers that prevent independent participation - Abuse through automated submissions, impersonation, or gate bypass - Using an educational instrument as professional, medical, financial, legal, safety, or engineering authority ## Qualification claim boundary ### Potentially allowed only after evidence and review - The supplied source record identifies a tool named Pointguard with tool_id pointguard and slug pointguard. - The source record lists Tilly as owner agent and lists Pointguard and independent reviewer as owner roles. - The source record associates Pointguard with independent review, instrument qualification, evidence reconciliation, grant research and drafts, and website claim review. - The source record lists Instrument packets, Qualification records, Grant dossiers, and Public-claim routes as related field guides. - The source record lists final publication, funding commitments, credential access, and changes to another agent's source as principal gates. - The source record prohibits approving one's own evidence, hiding a contradiction, and turning a draft into a public claim. - The source record cites agents/index.html#tilly. ### Still held - Pointguard is implemented, connected, deployed, or operational. - Pointguard has authority beyond the supplied owner, role, job, field-guide, gate, and never-rule statements. - Pointguard can approve final publication, funding commitments, credential access, or changes to another agent's source. - Pointguard has access to credentials, private evidence, external systems, or another agent's source. - Any instrument is qualified merely because it appears in the agent catalog. - Any grant, customer, deployment, effect, performance metric, provider, or current capability exists. - Any draft is a public claim. - Any qualification result is professional, medical, financial, legal, safety, or engineering approval. ## Sources and evidence These pointers bind the page to evidence. They do not grant authority or collapse verification into deployment. - `source_record:tool:pointguard` - `agents/index.html#tilly`