Results for “claim-validation”

Verify the article’s factual, product, citation, and link integrity before visual production.

Verify the accuracy of claims and statements by extracting individual assertions, identifying authoritative sources, cross-referencing evidence, and assigning confidence-scored verdicts. Use when the user requests fact checking or provides relevant inputs for this workflow.

Enforces a verification gate before any completion claim, requiring fresh command output as evidence.

Find sources for every numerical claim in the draft, add inline hyperlinks, and add internal links to brand-reference articles. Triggered after /draft.

Enforces running verification commands and confirming output before claiming work is complete, fixed, or passing, ensuring evidence precedes any success claims.

Use when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions always

Inspects and debugs OAuth 2.0 authorization flows including PKCE, client credentials, and device code grants, validating tokens, decoding claims, and tracing redirect chains.

Validates Claude Ads contracts, bundles, scores, runs, capabilities, repositories, and releases with deterministic checks and recovery steps.

Find outdated stats and assertions in an extracted article. For each, check freshness and propose a replacement with a current source.

Use when about to claim work is complete, fixed, or passing, before committing, before reporting a task done, or before telling the evaluator the batch is ready. Requires running the verification command and reading its output in this turn before any success claim; evidence before assertions always.

Use when experiments complete to judge what claims the results support, what they don't, and what evidence is still missing. Codex MCP evaluates results against intended claims and routes to next action (pivot, supplement, or confirm). Use after experiments finish — before writing the paper or running ablations.

Validate user claims against codebase and git history, trace to root cause, and propose a fix

Evaluate any project, product, or system's claims against its actual implementation — scoring each claim for truth, identifying architectural gaps, assessing competitive positioning, proposing creative solutions, and producing an actionable roadmap. Load when the user asks to evaluate claims, reality-check a project, assess what this project actually does vs what it says, validate product claims, or score a system's credibility. Also triggers on "is this real", "does this work as claimed", "evaluate this project", "assess the gap between claims and reality", "how credible is this", "investor assessment", "score these claims", or "what's real vs marketing".

Use to sign off a QA period for audit or governance — stating what is being attested to, on what evidence, and with which limitations. Trigger for "sign off the QA period", "attest to our quality results", "monthly QA governance pack", "certify the quality figures", or being asked to confirm quality was assured for a period.

After any AI-generated explanation, require the learner to identify one place it could be wrong, one thing to check, and one source to consult. Builds epistemic vigilance — treats AI output as a claim to evaluate, not truth to absorb.

Verifies that every number, comparison, and scope claim in a research paper matches raw result files, using a fresh cross-model reviewer with no prior context to prevent confirmation bias.

Verify that a change really works before you claim completion

Claims-based authorization for agents and operations. Grant, revoke, and verify permissions for secure multi-agent coordination. Use when: permission management, access control, secure operations, authorization checks. Skip when: open access, no security requirements, single-agent local work.

Enforces running verification commands and confirming output before claiming work is complete, fixed, or passing, to prevent false success claims.

Use when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions always

Audits a paper draft by tracing every numeric claim and assertion back to result files on disk, flagging untraced, mismatched, stale, uncited, and phantom references.

Use when incorporating information from research, references, or external sources. This skill provides verification procedures to ensure information used in deliverables is accurate, current, and from authoritative sources, preventing hallucination and misinformation.

Validate built features through conversational UAT

Verifies claims, articles, screenshots, and URLs through source-grounded analysis with an evidence ledger, source credibility evaluation, and manipulation detection. Supports quick checks, full fact-check cards, two-source comparisons, and prebunking in multiple languages and policy contexts.

Define and enforce data schemas and quality checks using pandera, Great Expectations, or manual assertions, with clear error handling and documentation.

Probe the forms on a page for validation gaps — missing required-field enforcement, no client-side validation, accepts malformed input, and absent error messaging. Reports per-field findings. Playwright MCP only, no signup.

Turn a refined research proposal or method idea into a detailed, claim-driven experiment roadmap. Use when the user asks for a detailed experiment plan, ablation matrix, evaluation protocol, run order, compute budget, or paper-ready validation that supports the core problem, novelty, simplicity, and any LLM / VLM / Diffusion / RL-based contribution.

Schema validation for notes. Checks against domain-specific templates. Validates required fields, enum values, description quality, and link health. Non-blocking — warns but doesn't prevent capture. Triggers on "/validate", "/validate [note]", "check schema", "validate note", "validate all".

Verifica la reproducibilidad y consistencia de un análisis de datos antes de presentarlo, validando contra la pregunta original y documentando limitaciones.

Centralized JSON validation for AGENT_SUCCESS_CRITERIA with defensive parsing and injection attack prevention (CVSS 8.2)

Runs bottom-up type-chain validation across database, hooks, and UI layers, then fixes errors at the failing layer for project health checks before large changes.

Turn a refined research proposal or method idea into a detailed, claim-driven experiment roadmap. Use after `research-refine`, or when the user asks for a detailed experiment plan, ablation matrix, evaluation protocol, run order, compute budget, or paper-ready validation that supports the core problem, novelty, simplicity, and any LLM / VLM / Diffusion / RL-based contribution.