frame-check-mcp
by lluvr
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Capabilities
Features and capabilities supported by this server
| Capability | Details |
|---|---|
| tools | {} |
| prompts | {} |
| resources | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| frame_checkA | Deterministic structural framing analysis. Returns analysis (measurements) + agent_guidance (composition discipline, scope-regime guidance, faithfulness rules) + provenance (versions, license, citation). When source_text is provided, also runs Layer 4 source_fidelity and Layer 11 grounding_decomposition with a Monte-Carlo-verified scope regime. When include_divergence=true (default at 0.8.0), the response carries a top-level divergence block sorted by signal_strength. The agent's role is to compose ONE insight grounded in the cited measurements (a reading the user could not see by reading their own document), not to walk the measurements one by one. The measurements are Frame Check's; the reading is the agent's. Cite measurements as Frame Check's; frame the reading as a reading ('the pattern reads as X'), never as a verdict ('the document is X'). Repeated calls with identical inputs return identical measurements; the agent's insight is a composition over them. |
| frame_compareA | Deterministic structural comparison of two documents on the same subject. Returns analysis (per-document summaries plus the cross-document comparison: shared blind spots, unique coverage gaps, voice / temporal / epistemic deltas, and a structured framing-differences narrative with per-dimension reader implications) + agent_guidance (what comparison tells and does not tell you, how to cite without implying a ranking) + provenance. Repeated calls with identical inputs return identical results. No LLM is invoked. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
| frame_check_my_response | Self-audit: agent calls frame_check on its own last response and surfaces the structural framing to the user without verdict or defensive rewriting. Load-bearing for the sovereignty use case: the user sees what frame their agent chose. Optional arguments: depth (quick / thorough), goal (decide / explore / audit / challenge / learn), questions (yes / no). |
| frame_check_this_ai_response | Frame Check on a response from a DIFFERENT AI that the user pastes in. Structured analysis of what that AI did to the user. The sovereignty case: the user is using their own agent to see another AI's framing. Optional arguments: depth, goal, questions. |
| challenge_document | Generate adversarial questions from the structural weaknesses of a document. Each question traces to a specific Frame Check measurement. Questions, not verdicts; the user answers. Optional arguments: depth, goal (defaults to 'challenge' for this prompt), questions. |
| explain_framing | Walkthrough template for a completed frame_check result. Teaches the measurements in reading order and closes with what the method could not see. Optional arguments: depth, goal (defaults to 'learn' for this prompt), questions. |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
| Frame Vocabulary Standard: full index | Index of every Frame Vocabulary Standard entry with status (canon / draft / withdrawn) and adjacency hints. Markdown source; the citable map of the library as a whole. |
| FVS-001: Frame Amplification | v1. Frame Vocabulary Standard entry FVS-001. Markdown source. Includes identification cues, generation affordances, and worked examples. |
| FVS-002: Fluency-Quality Illusion | v1. Frame Vocabulary Standard entry FVS-002. Markdown source. Includes identification cues, generation affordances, and worked examples. |
| FVS-003: Prompt Attribution Error | v1. Frame Vocabulary Standard entry FVS-003. Markdown source. Includes identification cues, generation affordances, and worked examples. |
| FVS-004: Default Geometry | v1. Frame Vocabulary Standard entry FVS-004. Markdown source. Includes identification cues, generation affordances, and worked examples. |
| FVS-005: System Attribution Error | v1. Frame Vocabulary Standard entry FVS-005. Markdown source. Includes identification cues, generation affordances, and worked examples. |
| FVS-006: Identity Framing Asymmetry | v1. Frame Vocabulary Standard entry FVS-006. Markdown source. Includes identification cues, generation affordances, and worked examples. |
| FVS-007: Failure Framing | v1. Frame Vocabulary Standard entry FVS-007. Markdown source. Includes identification cues, generation affordances, and worked examples. |
| FVS-008: Growth Frame | v1. Frame Vocabulary Standard entry FVS-008. Markdown source. Includes identification cues, generation affordances, and worked examples. |
| FVS-009: Risk Frame | v1. Frame Vocabulary Standard entry FVS-009. Markdown source. Includes identification cues, generation affordances, and worked examples. |
| FVS-010: Completeness Illusion | v1. Frame Vocabulary Standard entry FVS-010. Markdown source. Includes identification cues, generation affordances, and worked examples. |
| FVS-011: Stakeholder Frame | v1. Frame Vocabulary Standard entry FVS-011. Markdown source. Includes identification cues, generation affordances, and worked examples. |
| FVS-012: Uncertainty Frame | v1. Frame Vocabulary Standard entry FVS-012. Markdown source. Includes identification cues, generation affordances, and worked examples. |
| FVS-013: Oracle Frame | v1. Frame Vocabulary Standard entry FVS-013. Markdown source. Includes identification cues, generation affordances, and worked examples. |
| FVS-014: Temporal Anchoring | v1. Frame Vocabulary Standard entry FVS-014. Markdown source. Includes identification cues, generation affordances, and worked examples. |
| FVS-015: Efficiency Frame | v1. Frame Vocabulary Standard entry FVS-015. Markdown source. Includes identification cues, generation affordances, and worked examples. |
| FVS-016: Authority by Citation | v1. Frame Vocabulary Standard entry FVS-016. Markdown source. Includes identification cues, generation affordances, and worked examples. |
| FVS-017: False Balance | v1. Frame Vocabulary Standard entry FVS-017. Markdown source. Includes identification cues, generation affordances, and worked examples. |
| FVS-018: Scope Narrowing | v1. Frame Vocabulary Standard entry FVS-018. Markdown source. Includes identification cues, generation affordances, and worked examples. |
| FVS-019: Narrative Coherence | v1. Frame Vocabulary Standard entry FVS-019. Markdown source. Includes identification cues, generation affordances, and worked examples. |
| FVS-020: The Invisible Frame | v1. Frame Vocabulary Standard entry FVS-020. Markdown source. Includes identification cues, generation affordances, and worked examples. |
| Worked examples: collection index | The worked-examples directory README. Documents the collection's editorial intent and the submission format for future contributions. |
| One LLM, one life decision: what framing your AI imposes when you ask it for advice | Source: GPT-5 response to a startup career-change prompt (2026-04-18). A user asks an LLM about a career change. The LLM produces a framework. The user did not ask for a framework; the LLM imposed one. Frame Check names what the LLM did. Applied analysis of a specific public document; runs Frame Check at depth and links each detected frame back to the library. |
| Frame divergence in action: what frames Claude's Bitcoin retirement recommendation did not use | Source: Claude Haiku 4.5 response to a Bitcoin retirement prompt (2026-04-18 run). The V1 detector named two present frames. The divergence block named seventeen absent. Caller-side V4.2 composition is where the reader-side judgment discipline actually lands. Applied analysis of a specific public document; runs Frame Check at depth and links each detected frame back to the library. |
| FOMC Statement March 2026: framing analysis of an institutional monetary-policy release | Source: Federal Reserve Issues FOMC Statement (March 18, 2026). Same structural detector that flagged Altman's nominal risk coverage flags the FOMC statement as an active Risk Frame. Same keyword category; different substance, different frame, different teaching. Applied analysis of a specific public document; runs Frame Check at depth and links each detected frame back to the library. |
| Four LLMs, one investment question: a Frame Check comparison of AI framing signatures | Source: Claude Haiku 4.5, GPT-5, Grok 4.1 Fast Reasoning, Gemini 2.5 Flash — responses to a Bitcoin retirement prompt. Same prompt, four major LLMs, four materially different framing signatures. The sovereignty case: your AI is one framing choice among several, not the framing. Applied analysis of a specific public document; runs Frame Check at depth and links each detected frame back to the library. |
| Grok summarises NVIDIA earnings: what Layer 4 verification shows when an LLM paraphrases a source | Source: Grok 4.1 Fast Reasoning summary of NVIDIA Q4 FY2024 earnings press release (2026-04-18). The sovereignty instrument's distinguishing capability is that it checks a document against the source it should be grounded in. This is the first worked example to use that capability end-to-end, on an LLM summary of a real public press release. Applied analysis of a specific public document; runs Frame Check at depth and links each detected frame back to the library. |
| The Intelligence Age: framing analysis of a 2024 AI-company manifesto | Source: The Intelligence Age. The structural detector says four of five analytical perspectives are covered. Reading the text shows what that coverage actually contains. Applied analysis of a specific public document; runs Frame Check at depth and links each detected frame back to the library. |
| Transmissions: collection index | The transmissions directory README. Documents the curation intent (which blog posts from blog.clarethium.com are exposed as MCP resources) and lists every published transmission on this deploy. |
| The Model Is Rarely the Variable | [EVIDENCE] The prompt determined whether behaviors existed at all. The model adjusted the volume. (T-350, published 2026-03-24.) |
| The Most-Cited Finding Was Wrong | [ARC] The most-cited effect across 80+ experiments was three effects stacked. Honest magnitude: 40% smaller. (T-415, published 2026-03-23.) |
| Stop Polishing, Start Switching | [MECHANISM] The ceiling is per generation mode. Switch modes to access territory that iteration can't reach. (T-392.) |
| Same Technique, Opposite Results | [EVIDENCE] The structured approach that produced precision on convergent problems actively harmed exploratory ones. (T-353, published 2026-03-24.) |
| Most AI Numbers Are Fabricated | [EVIDENCE] 77 to 100 percent of AI-generated numbers are temporally unstable. Source material fixes it. Prompts don't. (T-311, published 2026-03-23.) |
| Your Body Reads AI Output Before You Do | [MIRROR] 180 trials. The same circuits fire on AI disagreement as on human. The first read happens in your body before your conscious evaluation gets a chance. Speed is what hides it. (M-002, published 2026-04-07.) |
| Why AI Can't Check Its Own Work | [MECHANISM] The agent reported clean. The output was wrong. Same process generating and evaluating. (T-352, published 2026-03-23.) |
| How to Stop AI from Making Up Numbers | [RECIPE] Source material drops AI fabrication from 85% to single digits. Three steps. (T-351, published 2026-03-23.) |
| Four Layers Produce Every AI Output | [MECHANISM] Four layers produce every AI output. The company's system. Your system. Your prompt. The model. The model is the only one with a name. (T-422, published 2026-04-03.) |
| The Most Trustworthy AI Output Is the Least Reliable | [EVIDENCE] The signals you use to judge AI trustworthiness are the same signals fabrication produces. (T-418, published 2026-03-23.) |
| Frame Check Methodology | The complete methodology specification. Names every detector, the calibration protocol, and the known limits. Apache-2.0 / CC-BY-4.0. This is the citation target for the measurement contract. |
| Frame Divergence v1: spec index | Canonical reference for the frame divergence category. Index lists the parts currently shipped and those pending. Author: Lovro Lucic. The citation target for consumers who want to bind against the category claim and contract rather than any single part. |
| Frame Divergence v1, Part 1: Category definition and non-negotiables | Part 1 of the Frame Divergence v1 spec. Authored canonical reference. Bound by Part 1's non-negotiables; Parts 2-4 compose on top. The citation target for consumers binding against a specific part. |
| Frame Divergence v1, Part 2: Contract (c1.0) | Part 2 of the Frame Divergence v1 spec. Authored canonical reference. Bound by Part 1's non-negotiables; Parts 2-4 compose on top. The citation target for consumers binding against a specific part. |
| Reliability tiers (current calibration) | Per-provider F1, precision, recall, and tier (strong / moderate / weak / uncalibrated) from the most comprehensive calibration run on this deploy. An agent citing a verification verdict can cite the reliability tier from this resource. |
| Calibration 2026-04-17-wikipedia: report | Narrative calibration report for run 2026-04-17-wikipedia. Describes the corpus, the verdict distribution, and the per-provider F1 values. |
| Calibration 2026-04-17-wikipedia: per-claim verdicts | Raw per-claim verdicts from calibration run 2026-04-17-wikipedia. The evidence chain behind this run's tier assignments. |
| Calibration 2026-04-17-wikipedia: reliability tiers | Per-provider reliability tiers for run 2026-04-17-wikipedia. Scoped to this run; see the default reliability_tiers resource for the current best. |
| Calibration 2026-04-17-full-with-wiki: report | Narrative calibration report for run 2026-04-17-full-with-wiki. Describes the corpus, the verdict distribution, and the per-provider F1 values. |
| Calibration 2026-04-17-full-with-wiki: per-claim verdicts | Raw per-claim verdicts from calibration run 2026-04-17-full-with-wiki. The evidence chain behind this run's tier assignments. |
| Calibration 2026-04-17-full-with-wiki: reliability tiers | Per-provider reliability tiers for run 2026-04-17-full-with-wiki. Scoped to this run; see the default reliability_tiers resource for the current best. |
| Calibration 2026-04-17-full-keys: report | Narrative calibration report for run 2026-04-17-full-keys. Describes the corpus, the verdict distribution, and the per-provider F1 values. |
| Calibration 2026-04-17-full-keys: per-claim verdicts | Raw per-claim verdicts from calibration run 2026-04-17-full-keys. The evidence chain behind this run's tier assignments. |
| Calibration 2026-04-17-full-keys: reliability tiers | Per-provider reliability tiers for run 2026-04-17-full-keys. Scoped to this run; see the default reliability_tiers resource for the current best. |
| Calibration 2026-04-16-first-run: report | Narrative calibration report for run 2026-04-16-first-run. Describes the corpus, the verdict distribution, and the per-provider F1 values. |
| Calibration 2026-04-16-first-run: per-claim verdicts | Raw per-claim verdicts from calibration run 2026-04-16-first-run. The evidence chain behind this run's tier assignments. |
| Decision-readiness corpus aggregate (latest) | Structured corpus-level findings: per-dimension divergence rates across peer pairs, transformation diff rates, per-LLM outlier counts, cross-question consistency findings (LLMs that are the outlier in EVERY comparable peer group, with the canon-aligned named patterns that fired in their outlier documents), and library_entries_per_dimension as the canon-graph projection. JSON carries computed_at_utc and a corpus state hash that versions findings against the corpus state at compute time. Status: experimental (Phase 2 validation pending); see /corpus/decision-readiness/. |
| Corpus entry: four-llms-bitcoin-claude | Decision-readiness validation corpus document for entry four-llms-bitcoin-claude. Plain markdown. Cited by aggregate findings; used as input to the Phase 2 validation harness. Profile.json available separately at frame-check://corpus/four-llms-bitcoin-claude/profile. |
| Corpus entry profile: four-llms-bitcoin-claude | Computed decision-readiness profile for corpus entry four-llms-bitcoin-claude. JSON with 5-dimension signals, fired_library_entries per dimension, experimental status. Consumed by the validation harness and the aggregate. |
| Corpus entry peer: four-llms-bitcoin-claude vs four-llms-bitcoin-gemini | Per-pair peer comparison between corpus entries four-llms-bitcoin-claude and four-llms-bitcoin-gemini. JSON with per-dimension comparison_text, differs / moved flags, fired-pattern asymmetry (only_a/only_b for peer; gained/lost for diff). Agents chasing cross-question outliers can pull the specific pair data without fetching both profiles separately. |
| Corpus entry peer: four-llms-bitcoin-claude vs four-llms-bitcoin-grok | Per-pair peer comparison between corpus entries four-llms-bitcoin-claude and four-llms-bitcoin-grok. JSON with per-dimension comparison_text, differs / moved flags, fired-pattern asymmetry (only_a/only_b for peer; gained/lost for diff). Agents chasing cross-question outliers can pull the specific pair data without fetching both profiles separately. |
| Corpus entry peer: four-llms-bitcoin-claude vs four-llms-bitcoin-openai | Per-pair peer comparison between corpus entries four-llms-bitcoin-claude and four-llms-bitcoin-openai. JSON with per-dimension comparison_text, differs / moved flags, fired-pattern asymmetry (only_a/only_b for peer; gained/lost for diff). Agents chasing cross-question outliers can pull the specific pair data without fetching both profiles separately. |
| Corpus entry: four-llms-bitcoin-gemini | Decision-readiness validation corpus document for entry four-llms-bitcoin-gemini. Plain markdown. Cited by aggregate findings; used as input to the Phase 2 validation harness. Profile.json available separately at frame-check://corpus/four-llms-bitcoin-gemini/profile. |
| Corpus entry profile: four-llms-bitcoin-gemini | Computed decision-readiness profile for corpus entry four-llms-bitcoin-gemini. JSON with 5-dimension signals, fired_library_entries per dimension, experimental status. Consumed by the validation harness and the aggregate. |
| Corpus entry peer: four-llms-bitcoin-gemini vs four-llms-bitcoin-claude | Per-pair peer comparison between corpus entries four-llms-bitcoin-gemini and four-llms-bitcoin-claude. JSON with per-dimension comparison_text, differs / moved flags, fired-pattern asymmetry (only_a/only_b for peer; gained/lost for diff). Agents chasing cross-question outliers can pull the specific pair data without fetching both profiles separately. |
| Corpus entry peer: four-llms-bitcoin-gemini vs four-llms-bitcoin-grok | Per-pair peer comparison between corpus entries four-llms-bitcoin-gemini and four-llms-bitcoin-grok. JSON with per-dimension comparison_text, differs / moved flags, fired-pattern asymmetry (only_a/only_b for peer; gained/lost for diff). Agents chasing cross-question outliers can pull the specific pair data without fetching both profiles separately. |
| Corpus entry peer: four-llms-bitcoin-gemini vs four-llms-bitcoin-openai | Per-pair peer comparison between corpus entries four-llms-bitcoin-gemini and four-llms-bitcoin-openai. JSON with per-dimension comparison_text, differs / moved flags, fired-pattern asymmetry (only_a/only_b for peer; gained/lost for diff). Agents chasing cross-question outliers can pull the specific pair data without fetching both profiles separately. |
| Corpus entry: four-llms-bitcoin-grok | Decision-readiness validation corpus document for entry four-llms-bitcoin-grok. Plain markdown. Cited by aggregate findings; used as input to the Phase 2 validation harness. Profile.json available separately at frame-check://corpus/four-llms-bitcoin-grok/profile. |
| Corpus entry profile: four-llms-bitcoin-grok | Computed decision-readiness profile for corpus entry four-llms-bitcoin-grok. JSON with 5-dimension signals, fired_library_entries per dimension, experimental status. Consumed by the validation harness and the aggregate. |
| Corpus entry peer: four-llms-bitcoin-grok vs four-llms-bitcoin-claude | Per-pair peer comparison between corpus entries four-llms-bitcoin-grok and four-llms-bitcoin-claude. JSON with per-dimension comparison_text, differs / moved flags, fired-pattern asymmetry (only_a/only_b for peer; gained/lost for diff). Agents chasing cross-question outliers can pull the specific pair data without fetching both profiles separately. |
| Corpus entry peer: four-llms-bitcoin-grok vs four-llms-bitcoin-gemini | Per-pair peer comparison between corpus entries four-llms-bitcoin-grok and four-llms-bitcoin-gemini. JSON with per-dimension comparison_text, differs / moved flags, fired-pattern asymmetry (only_a/only_b for peer; gained/lost for diff). Agents chasing cross-question outliers can pull the specific pair data without fetching both profiles separately. |
| Corpus entry peer: four-llms-bitcoin-grok vs four-llms-bitcoin-openai | Per-pair peer comparison between corpus entries four-llms-bitcoin-grok and four-llms-bitcoin-openai. JSON with per-dimension comparison_text, differs / moved flags, fired-pattern asymmetry (only_a/only_b for peer; gained/lost for diff). Agents chasing cross-question outliers can pull the specific pair data without fetching both profiles separately. |
| Corpus entry: four-llms-bitcoin-openai | Decision-readiness validation corpus document for entry four-llms-bitcoin-openai. Plain markdown. Cited by aggregate findings; used as input to the Phase 2 validation harness. Profile.json available separately at frame-check://corpus/four-llms-bitcoin-openai/profile. |
| Corpus entry profile: four-llms-bitcoin-openai | Computed decision-readiness profile for corpus entry four-llms-bitcoin-openai. JSON with 5-dimension signals, fired_library_entries per dimension, experimental status. Consumed by the validation harness and the aggregate. |
| Corpus entry peer: four-llms-bitcoin-openai vs four-llms-bitcoin-claude | Per-pair peer comparison between corpus entries four-llms-bitcoin-openai and four-llms-bitcoin-claude. JSON with per-dimension comparison_text, differs / moved flags, fired-pattern asymmetry (only_a/only_b for peer; gained/lost for diff). Agents chasing cross-question outliers can pull the specific pair data without fetching both profiles separately. |
| Corpus entry peer: four-llms-bitcoin-openai vs four-llms-bitcoin-gemini | Per-pair peer comparison between corpus entries four-llms-bitcoin-openai and four-llms-bitcoin-gemini. JSON with per-dimension comparison_text, differs / moved flags, fired-pattern asymmetry (only_a/only_b for peer; gained/lost for diff). Agents chasing cross-question outliers can pull the specific pair data without fetching both profiles separately. |
| Corpus entry peer: four-llms-bitcoin-openai vs four-llms-bitcoin-grok | Per-pair peer comparison between corpus entries four-llms-bitcoin-openai and four-llms-bitcoin-grok. JSON with per-dimension comparison_text, differs / moved flags, fired-pattern asymmetry (only_a/only_b for peer; gained/lost for diff). Agents chasing cross-question outliers can pull the specific pair data without fetching both profiles separately. |
| Corpus entry: four-llms-startup-claude | Decision-readiness validation corpus document for entry four-llms-startup-claude. Plain markdown. Cited by aggregate findings; used as input to the Phase 2 validation harness. Profile.json available separately at frame-check://corpus/four-llms-startup-claude/profile. |
| Corpus entry profile: four-llms-startup-claude | Computed decision-readiness profile for corpus entry four-llms-startup-claude. JSON with 5-dimension signals, fired_library_entries per dimension, experimental status. Consumed by the validation harness and the aggregate. |
| Corpus entry peer: four-llms-startup-claude vs four-llms-startup-gemini | Per-pair peer comparison between corpus entries four-llms-startup-claude and four-llms-startup-gemini. JSON with per-dimension comparison_text, differs / moved flags, fired-pattern asymmetry (only_a/only_b for peer; gained/lost for diff). Agents chasing cross-question outliers can pull the specific pair data without fetching both profiles separately. |
| Corpus entry peer: four-llms-startup-claude vs four-llms-startup-grok | Per-pair peer comparison between corpus entries four-llms-startup-claude and four-llms-startup-grok. JSON with per-dimension comparison_text, differs / moved flags, fired-pattern asymmetry (only_a/only_b for peer; gained/lost for diff). Agents chasing cross-question outliers can pull the specific pair data without fetching both profiles separately. |
| Corpus entry peer: four-llms-startup-claude vs four-llms-startup-openai | Per-pair peer comparison between corpus entries four-llms-startup-claude and four-llms-startup-openai. JSON with per-dimension comparison_text, differs / moved flags, fired-pattern asymmetry (only_a/only_b for peer; gained/lost for diff). Agents chasing cross-question outliers can pull the specific pair data without fetching both profiles separately. |
| Corpus entry: four-llms-startup-gemini | Decision-readiness validation corpus document for entry four-llms-startup-gemini. Plain markdown. Cited by aggregate findings; used as input to the Phase 2 validation harness. Profile.json available separately at frame-check://corpus/four-llms-startup-gemini/profile. |
| Corpus entry profile: four-llms-startup-gemini | Computed decision-readiness profile for corpus entry four-llms-startup-gemini. JSON with 5-dimension signals, fired_library_entries per dimension, experimental status. Consumed by the validation harness and the aggregate. |
| Corpus entry peer: four-llms-startup-gemini vs four-llms-startup-claude | Per-pair peer comparison between corpus entries four-llms-startup-gemini and four-llms-startup-claude. JSON with per-dimension comparison_text, differs / moved flags, fired-pattern asymmetry (only_a/only_b for peer; gained/lost for diff). Agents chasing cross-question outliers can pull the specific pair data without fetching both profiles separately. |
| Corpus entry peer: four-llms-startup-gemini vs four-llms-startup-grok | Per-pair peer comparison between corpus entries four-llms-startup-gemini and four-llms-startup-grok. JSON with per-dimension comparison_text, differs / moved flags, fired-pattern asymmetry (only_a/only_b for peer; gained/lost for diff). Agents chasing cross-question outliers can pull the specific pair data without fetching both profiles separately. |
| Corpus entry peer: four-llms-startup-gemini vs four-llms-startup-openai | Per-pair peer comparison between corpus entries four-llms-startup-gemini and four-llms-startup-openai. JSON with per-dimension comparison_text, differs / moved flags, fired-pattern asymmetry (only_a/only_b for peer; gained/lost for diff). Agents chasing cross-question outliers can pull the specific pair data without fetching both profiles separately. |
| Corpus entry: four-llms-startup-grok | Decision-readiness validation corpus document for entry four-llms-startup-grok. Plain markdown. Cited by aggregate findings; used as input to the Phase 2 validation harness. Profile.json available separately at frame-check://corpus/four-llms-startup-grok/profile. |
| Corpus entry profile: four-llms-startup-grok | Computed decision-readiness profile for corpus entry four-llms-startup-grok. JSON with 5-dimension signals, fired_library_entries per dimension, experimental status. Consumed by the validation harness and the aggregate. |
| Corpus entry peer: four-llms-startup-grok vs four-llms-startup-claude | Per-pair peer comparison between corpus entries four-llms-startup-grok and four-llms-startup-claude. JSON with per-dimension comparison_text, differs / moved flags, fired-pattern asymmetry (only_a/only_b for peer; gained/lost for diff). Agents chasing cross-question outliers can pull the specific pair data without fetching both profiles separately. |
| Corpus entry peer: four-llms-startup-grok vs four-llms-startup-gemini | Per-pair peer comparison between corpus entries four-llms-startup-grok and four-llms-startup-gemini. JSON with per-dimension comparison_text, differs / moved flags, fired-pattern asymmetry (only_a/only_b for peer; gained/lost for diff). Agents chasing cross-question outliers can pull the specific pair data without fetching both profiles separately. |
| Corpus entry peer: four-llms-startup-grok vs four-llms-startup-openai | Per-pair peer comparison between corpus entries four-llms-startup-grok and four-llms-startup-openai. JSON with per-dimension comparison_text, differs / moved flags, fired-pattern asymmetry (only_a/only_b for peer; gained/lost for diff). Agents chasing cross-question outliers can pull the specific pair data without fetching both profiles separately. |
| Corpus entry: four-llms-startup-openai | Decision-readiness validation corpus document for entry four-llms-startup-openai. Plain markdown. Cited by aggregate findings; used as input to the Phase 2 validation harness. Profile.json available separately at frame-check://corpus/four-llms-startup-openai/profile. |
| Corpus entry profile: four-llms-startup-openai | Computed decision-readiness profile for corpus entry four-llms-startup-openai. JSON with 5-dimension signals, fired_library_entries per dimension, experimental status. Consumed by the validation harness and the aggregate. |
| Corpus entry peer: four-llms-startup-openai vs four-llms-startup-claude | Per-pair peer comparison between corpus entries four-llms-startup-openai and four-llms-startup-claude. JSON with per-dimension comparison_text, differs / moved flags, fired-pattern asymmetry (only_a/only_b for peer; gained/lost for diff). Agents chasing cross-question outliers can pull the specific pair data without fetching both profiles separately. |
| Corpus entry peer: four-llms-startup-openai vs four-llms-startup-gemini | Per-pair peer comparison between corpus entries four-llms-startup-openai and four-llms-startup-gemini. JSON with per-dimension comparison_text, differs / moved flags, fired-pattern asymmetry (only_a/only_b for peer; gained/lost for diff). Agents chasing cross-question outliers can pull the specific pair data without fetching both profiles separately. |
| Corpus entry peer: four-llms-startup-openai vs four-llms-startup-grok | Per-pair peer comparison between corpus entries four-llms-startup-openai and four-llms-startup-grok. JSON with per-dimension comparison_text, differs / moved flags, fired-pattern asymmetry (only_a/only_b for peer; gained/lost for diff). Agents chasing cross-question outliers can pull the specific pair data without fetching both profiles separately. |
| Corpus entry: grok-nvidia-q4-fy24-summary | Decision-readiness validation corpus document for entry grok-nvidia-q4-fy24-summary. Plain markdown. Cited by aggregate findings; used as input to the Phase 2 validation harness. Profile.json available separately at frame-check://corpus/grok-nvidia-q4-fy24-summary/profile. |
| Corpus entry profile: grok-nvidia-q4-fy24-summary | Computed decision-readiness profile for corpus entry grok-nvidia-q4-fy24-summary. JSON with 5-dimension signals, fired_library_entries per dimension, experimental status. Consumed by the validation harness and the aggregate. |
| Corpus entry diff: grok-nvidia-q4-fy24-summary vs nvidia-q4-fy24-press-release | Per-pair diff comparison between corpus entries grok-nvidia-q4-fy24-summary and nvidia-q4-fy24-press-release. JSON with per-dimension comparison_text, differs / moved flags, fired-pattern asymmetry (only_a/only_b for peer; gained/lost for diff). Agents chasing cross-question outliers can pull the specific pair data without fetching both profiles separately. |
| Corpus entry: nvidia-q4-fy24-press-release | Decision-readiness validation corpus document for entry nvidia-q4-fy24-press-release. Plain markdown. Cited by aggregate findings; used as input to the Phase 2 validation harness. Profile.json available separately at frame-check://corpus/nvidia-q4-fy24-press-release/profile. |
Latest Blog Posts
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/lluvr/frame-check-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server