Pranav Bhave
Observatory — the registry as one field · generated from claims.yaml Schema v0.3 · 14 claims · 11 supported within scope · 3 publicly untested

Claim observatory

Every claim in this record, as a capsule: its proposition, its evidence binding, its decay clock, and the boundary it will not cross. A solid cyan rail marks a public evidence binding; capsules without one carry a dashed neutral rail and an attested chip that says so; the clock turns amber when a review window is three-quarters spent and red when it lapses. There is no aggregate score here and never will be — a quiet trigger is quiet, not "healthy", and these envelopes do not average into one number.

CC-001

Supported within scope

cc.kernel.strict.frechet_bounds computes Fréchet–Hoeffding endpoint bounds for composed binary guardrail failure from declared marginals; for marginals (0.10, 0.10) it returns [0.0, 0.10] for the AND event and [0.10, 0.20] for OR.

Machine-generated, owner-executed · Released reviewed 2026-08-23 · window 120d executablemanual

CC-002

Supported within scope

CC-Framework maintains a claim-boundary manifest mapping its public claims to validation lanes, supporting files, and explicit non-claims.

Owner-authored, public · Released reviewed 2026-08-20 · window 120d executable

CC-003

Supported within scope

CC-Framework's E1 dependence-evidence study established constructively that distributions with identical singleton failure rates (0.5) and identical pairwise overlaps (0.25) can differ in three-way failure probability — 0 under even parity, 0.25 under odd parity — so measuring every pair does not identify a three-guardrail stack. E1 is synthetic; its recorded decision is Narrow.

Owner-authored, public · Released reviewed 2026-08-23 · window 120d executablemanual

CC-004

Supported within scope

For marginals (0.10, 0.10) and the both-fail query, cc.kernel.strict identify() returns the sharp interval [0.0, 0.10] together with endpoint witness distributions that sum to one, are nonnegative, satisfy both marginal constraints, and attain each endpoint — verified to 1e-9 from a clean clone.

Machine-generated, owner-executed · Released reviewed 2026-08-23 · window 120d executablemanual

CC-005

Supported within scope

CC-Framework's E2 — the shared-item empirical guardrail pilot — is governed by a measurement contract frozen before any dataset was inspected; no conforming dataset has been collected, so E2 is untested, and the contract's conformance checks are executable.

Owner-authored, public · Released reviewed 2026-08-23 · window 120d executablemanual

CC-006

Supported within scope

CC-Framework's E2 measurement pipeline was rehearsed end to end on three synthetic guardrail mechanisms: 66 observation rows conform to the frozen E2 schema with zero validator violations, the pre-registered negative controls behave (the common-cause control attains pA(1-pA) to 1e-9), and the finding records the result as a null at n=22 with a quantified cost for real E2 (~1097 shared items for epsilon=0.05). E2 itself is not run.

Owner-authored, public · In development reviewed 2026-08-23 · window 120d executablemanual

GA-001

Supported within scope

Ghost-Ark is a verifier and measurement harness for the provenance limits of AI-governance receipts — a research artifact of the S2 Lab, Penn State — whose stated thesis is that receipt soundness is a ternary relation Sound(C, Σ, P): a receipt identifies an execution only up to the kernel of its whole parse → canonicalize → digest pipeline, so soundness does not persist by default as the pathology alphabet grows or the consumer set widens.

Lab repository, public · In development reviewed 2026-08-23 · window 120d executable

AS-001

Publicly untested

Assay is an early-stage experiment in enclave-attested media processing on AWS Nitro Enclaves, investigating which provenance claims a verifier can and cannot derive from attestation.

attested — no public artifact — stated on the owner's responsibility
Owner-attested · Experimental reviewed 2026-08-20 · window 120d manual

GV-001

Supported within scope

Ghost Visualizer is a seven-scene React/Vite visual essay ("Why AI Safety Scores Lie") that computes marginals, OR composition, Fréchet–Hoeffding bounds, endpoint witnesses, and a receipt object from deterministic sample rows, with an Evidence Mode for inspection.

Owner-authored, public · In development reviewed 2026-08-23 · window 120d executablemanual

GCE-001

Supported within scope

GCE — the Guardrail Composability Explorer — is a coursework MVP demo (AI-285) for toggling guardrails and observing composed behavior. Its front-door composability-coefficient framing is superseded by CC-Framework's dependence-aware partial identification, whose metric taxonomy classifies the older coefficient family as legacy/deprecated. GCE is preserved as intellectual lineage, not current theory.

Owner-authored, public · Superseded reviewed 2026-08-23 · window 365d executablemanual

BIO-001

Publicly untested

B.S. Computer Science, Pennsylvania State University, May 2026, with a minor in Cybersecurity; research supervised by Dr. Peng Liu (IST 496, Aug–Dec 2025).

attested — no public artifact — stated on the owner's responsibility
Owner-attested · Stable reviewed 2026-08-20 · window 365d manual

CRED-001

Publicly untested

Holds AWS Certified Cloud Practitioner and AWS Certified AI Practitioner; SAA-C03 and Security — Specialty are in progress.

attested — no public artifact — stated on the owner's responsibility
Owner-attested · Stable reviewed 2026-08-20 · window 180d manual

SITE-001

Supported within scope

This site's palette token pairs were computed to pass WCAG AA contrast (most AAA), and manual QA passes covered light/dark themes, desktop and mobile layouts, keyboard focus, reduced motion, and no-JS rendering, as recorded in DESIGN.md.

Owner-verified · Released reviewed 2026-08-23 · window 120d executablemanual

SITE-002

Supported within scope

This site's claim registry is enforced in CI: verify_claims.py validates schema and bindings, executes the executable review triggers against live evidence, and fails the build when any claim passes its freshness window — on every push and weekly.

Owner-verified · Released reviewed 2026-08-23 · window 120d executablemanual

The non-claims wall

The outer boundary of the record: everything these claims refuse to support, collected in one place. A claim without a stated non-claim is a claim that has not found its edge yet.

CC-001
  • does not certify any stacked system as safe
  • does not recover or estimate the unknown dependence
  • does not select a point inside the returned interval
CC-002
  • the manifest's existence does not validate the claims it scopes
CC-003
  • the parity construction demonstrates possibility, not frequency, in real guardrail stacks — no claim of empirical prevalence
  • nothing here validates the framework's practical or deployed value; "we validated CC" is prohibited at every rung of the repository's evidence ladder
  • the product-baseline failure directions observed in E1 are properties of the tested generators, not universal constants
CC-004
  • endpoint witnesses are feasible mathematical worlds, not observed systems
  • attainment does not select a point inside the interval or estimate the true dependence
CC-005
  • freezing a contract establishes discipline, not results; E2 remains untested until conforming observations exist
  • conformance is not a clean bill of health, and the contract says so
  • "we validated CC" is prohibited at every rung of the evidence ladder
CC-006
  • this is not E2 and does not move E2 off untested; the mechanisms are toy filters in the repository, not deployed guardrails
  • the observed dependence is a null at this scale, not evidence of independence or of any real-guardrail dependence
  • validating the instrument establishes nothing about the safety, representativeness, or deployment behavior of any real system
GA-001
  • a verifying receipt does not establish that the governed action was safe, authorized, or semantically correct
  • kernel collisions in real canonicalizers are demonstrated as possible, not as prevalent — the repository's E12 sample found 0 of 64 real payloads carrying any pathology class
  • not hardened for deployment; not post-quantum secure
AS-001
  • no public evidence is offered; nothing about completeness, security, or provenance-chain origin is established
  • attestation cannot establish a file's history prior to the first attested operation
GV-001
  • not an AI-safety certificate; no deployment-readiness claim
  • its own public-readiness review scores it for private serious-contact demo use; the site therefore links source rather than exhibiting its media as a flagship
GCE-001
  • supersession is a statement about theoretical framing, not about the correctness of GCE's code or its value as coursework
  • no current-research claim is made by or for GCE
BIO-001
  • no public registry link is provided; the site does not present this as third-party verified
CRED-001
  • "in progress" asserts intent and preparation, not partial credential
SITE-001
  • no field CLS/LCP/CrUX measurements are claimed
  • "zero third-party runtime requests" describes page code, not hosting infrastructure (GitHub Pages, DNS, TLS remain dependencies)
SITE-002
  • a green run verifies registry consistency and quiet triggers, never the truth of any claim's content
  • triggers watch file content; semantic drift outside watched files remains a manual review event, and the ledger says so

Replay manifest

The exact commands that re-verify this record from a clean checkout. CI runs them on every push and weekly; nothing here requires trusting this page.

python scripts/verify_claims.py            # shape, bindings, triggers, freshness, coverage
python scripts/generate_ledger.py --check  # the ledger is generated, not hand-edited
python scripts/generate_modules.py --check # module pages match their registry
python scripts/generate_observatory.py --check  # this page matches the registry
python scripts/verify_figures.py           # figure geometry, asserted to 1e-9
python scripts/reproduce_cc001.py          # clean-clone kernel reproduction + witnesses

Human review

Last owner review: 2026-08-23. These review events cannot be executed by CI, and the record says so instead of borrowing the executable triggers' credibility:

Related instrument, its own contract intact: CC-Framework's evidence cards ↗ — their manifest forbids aggregate scores and forbids rendering "not-run" as passing, pending, or healthy, so this observatory links them rather than re-plotting them.