Skip to content

docs(readme): reframe PSE as post-symbolic cognitive substrate#85

Merged
LashSesh merged 1 commit into
mainfrom
claude/add-external-fixture-loader-TAp7N
May 17, 2026
Merged

docs(readme): reframe PSE as post-symbolic cognitive substrate#85
LashSesh merged 1 commit into
mainfrom
claude/add-external-fixture-loader-TAp7N

Conversation

@LashSesh
Copy link
Copy Markdown
Owner

The previous README described PSE in terms of its original data-analysis pipeline identity ("streaming computation engine that detects structural events in observation streams"). This was accurate for the core engine before the traversal stack was built, but no longer reflects what the system is.

Rewrite to accurately reflect the current architecture:

  • Opening: PSE is a post-symbolic cognitive substrate, not a stream detector. It sits between symbolic and connectionist computation, reducing to neither.
  • Architecture section: three-tier model (Core Engine / Traversal Agent Stack / Governance) with layer table and ADAMANT protocol explained.
  • Layer detail sections: each layer gets its own section with the correct pipeline diagram and invariants (previously only described in a "What's new" changelog block at the bottom).
  • Status table: added ADAMANT v1.0.0 and productive-task validation as explicit rows.
  • Eval matrix status: surface the diagnostic results (PSE 100% recall, Raw 52-63%) and the productive_agent_validated=false status honestly.
  • Specifications table: replaces scattered inline references with a single structured table mapping each document to its layer.
  • Removed: the "What's new since the last README" changelog section (content belongs in CHANGELOG.md, not the primary README).

No technical content was removed; all CLI commands, benchmark numbers, crate descriptions, and quality/CI tables are preserved.

https://claude.ai/code/session_01K5AN3s9TnGo1Az4jYwagtw

The previous README described PSE in terms of its original data-analysis
pipeline identity ("streaming computation engine that detects structural
events in observation streams"). This was accurate for the core engine
before the traversal stack was built, but no longer reflects what the
system is.

Rewrite to accurately reflect the current architecture:

- Opening: PSE is a post-symbolic cognitive substrate, not a stream
  detector. It sits between symbolic and connectionist computation,
  reducing to neither.
- Architecture section: three-tier model (Core Engine / Traversal Agent
  Stack / Governance) with layer table and ADAMANT protocol explained.
- Layer detail sections: each layer gets its own section with the
  correct pipeline diagram and invariants (previously only described
  in a "What's new" changelog block at the bottom).
- Status table: added ADAMANT v1.0.0 and productive-task validation
  as explicit rows.
- Eval matrix status: surface the diagnostic results (PSE 100% recall,
  Raw 52-63%) and the productive_agent_validated=false status honestly.
- Specifications table: replaces scattered inline references with a
  single structured table mapping each document to its layer.
- Removed: the "What's new since the last README" changelog section
  (content belongs in CHANGELOG.md, not the primary README).

No technical content was removed; all CLI commands, benchmark numbers,
crate descriptions, and quality/CI tables are preserved.

https://claude.ai/code/session_01K5AN3s9TnGo1Az4jYwagtw
@LashSesh LashSesh merged commit 1dcd7b5 into main May 17, 2026
@LashSesh LashSesh deleted the claude/add-external-fixture-loader-TAp7N branch May 17, 2026 17:24
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants