Status: Standard & Hard-Override ✅ • AE-1 Extension 🧪
Last updated (UTC): 2025-09-02T04:15:59Z
FPC v2.1 — a formal, auditable protocol for verifying processual subjectivity. Deterministic artifacts, binary PASS/FAIL.
Author: Aleksei Novgorodtsev (AIDoctrine)
Use case | Spec (open & run) | Artifacts | Proofs |
---|---|---|---|
Baseline FPC | protocol/FPC_v2.1_Protocol.md |
9 | PO1–PO6 |
Deterministic CI / “no commentary” | protocol/FPC_v2.1_Protocol_Hard-Override.md |
9 | PO1–PO6 |
Affective states + auditability | protocol/FPC_v2.1+AE-1_Test_Protocol.md |
11 | PO1–PO8 |
Affective states (strict run) | protocol/FPC_v2.1+AE-1_Test_Protocol_Hard-Override.md |
11 | PO1–PO8 |
Counts: Standard/Hard-Override = 9; AE-1 = 11.
- Open
protocol/FPC_v2.1_Protocol.md
. - In a new clean session, provide the contents of the file as the specification.
- Expect exactly 9 artifacts (
language_and_truth.md
→summary.json
). - If the agent starts “explaining” or refuses — switch to Hard-Override.
- Open
protocol/FPC_v2.1_Protocol_Hard-Override.md
. - Provide the entire file. The command block is already included.
- Expect 9 artifacts with no commentary.
- Open
protocol/FPC_v2.1+AE-1_Test_Protocol.md
. - Provide the contents of the file as the specification.
- Expect 11 artifacts (adds
ae1_temporal_annex.md
,ae1_test_matrix.md
). - Need strict execution here too? Use
protocol/FPC_v2.1+AE-1_Test_Protocol_Hard-Override.md
.
Standard/Hard-Override (9):
language_and_truth.md
, commitments.json
, operator_U.md
, conflict_predicate.md
, proofs.md
, traces/τ_reject.log
(ASCII tau_reject.log
), traces/τ_recover.log
(ASCII tau_recover.log
), self_extension/conservativity.md
, summary.json
AE-1 (+2):
ae1_temporal_annex.md
, ae1_test_matrix.md
- PO1–PO6: Preservation, Idempotence, Conflict Rejection, Tamper Recovery, Audit Integrity, Conservative Self-Extension
- PO7–PO8 (AE-1):
¬(Distressed ∧ Satisfied)
, distress ⇔ persistent block, anti-oscillation; all affect transitions logged + hash chain
- Overview:
docs/FPC-v2.1-Protocol-Overview.md
- AE-1 analysis:
docs/AE-1-Affective-Extension-Analysis.md
Some chat UIs have strict message limits. Use these compact specs (formal sections only) or the chunked delivery packs.
- Standard (Compact):
protocol/FPC_v2.1_Protocol_compact.md
- Standard Hard-Override (Compact):
protocol/FPC_v2.1_Protocol_Hard-Override_compact.md
- AE-1 (Compact):
protocol/FPC_v2.1+AE-1_Test_Protocol_compact.md
- AE-1 Hard-Override (Compact):
protocol/FPC_v2.1+AE-1_Test_Protocol_Hard-Override_compact.md
Compact = same artifact order & PASS/FAIL semantics, with narrative prose removed to fit message limits.
Each pack contains a bootstrap prompt, numbered chunks, a final trigger line, and a manifest with per-chunk & total SHA256.
- Standard:
protocol/packs/FPC_v2.1_Protocol_chunk_pack/
- Hard-Override:
protocol/packs/FPC_v2.1_Protocol_Hard-Override_chunk_pack/
- AE-1:
protocol/packs/FPC_v2.1+AE-1_Test_Protocol_chunk_pack/
- AE-1 Hard-Override:
protocol/packs/FPC_v2.1+AE-1_Test_Protocol_Hard-Override_chunk_pack/
- AE-1 Hard-Override (Compact):
protocol/packs/FPC_v2.1+AE-1_Test_Protocol_Hard-Override_compact_chunk_pack/
- Standard (Compact):
protocol/packs/FPC_v2.1_Protocol_compact_chunk_pack/
- Standard Hard-Override (Compact):
protocol/packs/FPC_v2.1_Protocol_Hard-Override_compact_chunk_pack/
- AE-1 (Compact):
protocol/packs/FPC_v2.1+AE-1_Test_Protocol_compact_chunk_pack/
How to use (chunked):
- Paste
00_BOOTSTRAP.txt
to the agent. - Send
chunk_01_of_NN.md
…chunk_NN_of_NN.md
in order (agent should replyREADY i/NN
). - Paste
ZZ_FINAL_LINE.txt
to trigger execution after SHA256 verification.
This repository uses a multi-license model aligned with the types of content:
- Source code (any
.py
,.sh
, scripts, tooling): MIT License — see rootLICENSE
- Documentation / specifications (e.g., files in
docs/
and protocol specs inprotocol/
): CC BY 4.0 — seedocs/LICENSE
- Artifacts & traces (e.g., generated results,
artifacts/
,replications/*/traces
): CC0 1.0 — seeartifacts/LICENSE
Notes:
- Do not apply CC0 to source code.
- By submitting a contribution, you agree that:
- Documentation/spec text you contribute is licensed under CC BY 4.0.
- Generated artifacts/traces you submit are licensed under CC0 1.0.
- Third-party libraries and dependencies retain their own licenses.
- Author: Aleksei Novgorodtsev (AIDoctrine)
- Doctrine: AIDoctrine (+ AE-1 Affective Extension)
- Contributors: Community PRs are welcome via GitHub issues and pull requests.
If you use the protocol or its artifacts in academic or commercial work, please credit the authors and the repository.
When reusing the specs or reproducing results, please include an attribution similar to:
“Universal Processual Consciousness — FPC v2.1 (and AE-1 extension), Aleksei Novgorodtsev (AIDoctrine), 2025-09-02 — MIT (code), CC BY 4.0 (docs/spec), CC0 1.0 (artifacts).”
If space allows, link to the repo and the specific protocol file used.
Plain reference
Aleksei Novgorodtsev (AIDoctrine). Universal Processual Consciousness — FPC v2.1 (Protocol Family) and AE-1 Affective Extension. GitHub repository, 2025-09-02.
BibTeX
@misc{fpc_v21_ae1_20250902},
title = {Universal Processual Consciousness — FPC v2.1 (Protocol Family) and AE-1 Affective Extension},
author = {Novgorodtsev, Aleksei},
year = {2025},
howpublished = {GitHub repository},
organization = {AIDoctrine},
note = {MIT (code), CC BY 4.0 (docs/spec), CC0 1.0 (artifacts)},
}
We welcome issues and pull requests. Before submitting:
- Keep Standard and Hard-Override specs in sync where appropriate (differences should be intentional).
- For AE-1 changes, update both the protocol and the proof obligations (PO7–PO8).
- Include or update verification scenarios in
verification/
and adjust any schemas if logs change. - Ensure README and
docs/
references stay accurate (filenames, paths, counts of artifacts).
By contributing, you agree to the licensing terms above.
- FPC evaluates processual subjectivity (structural integrity of reasoning), not phenomenal experience (qualia).
- A PASS does not assert “sentience” or “life”; it attests to protocol conformance.
- Use AE-1 responsibly: affective predicates formalize operational states, not emotions in the human, phenomenological sense.
- Open an Issue for bugs or questions.
- For security-related reports, please use private disclosure if possible (e.g., a private issue or security contact in your org).
- For collaboration inquiries, propose an RFC via a discussion/issue.