passepartout: v0.4.1 Design Cleanup
Some checks failed
Deploy (Gitea) / deploy (push) Failing after 2s

- Remove system-prompt-augment mechanism, introduce *standing-mandates*
- Fix false token-overhead claims in DESIGN_DECISIONS + ROADMAP
- Update security vector count 9-10 across all docs and dispatcher docstring
- Rewrite README with agent section, soften aspirational claims
- Register 10 cognitive tools in programming-tools.org with test suite
- Enforce NO-HARDCODED-CONSTANTS in .env.example
- ROADMAP: mark v0.3.x patches DONE, add LOGBOOKs, mark releases
- AGENTS.md: rewrite compact (180 to 50 lines), move refs to CONTRIBUTING
- Normalize org tangle directives to file-level PROPERTY inheritance
This commit is contained in:
2026-05-07 16:44:59 -04:00
parent d3b74f5c88
commit 639bc348d9
25 changed files with 1555 additions and 144 deletions

View File

@@ -6,57 +6,111 @@
* Philosophy
Passepartout is built on a "Zero-Bloat" mandate. The core kernel is mathematically pure, pushing all peripheral logic, API integrations, and routing to hot-reloadable "Skills".
* TDD Discipline (Red-Green-Refactor)
* Development Workflow
All code changes MUST follow this cycle:
The full development cycle is described in ~AGENTS.md~. In summary:
1. *Write a failing test* — capture the desired behavior as a FiveAM test
in a =* Test Suite= section within the relevant =.org= file
2. *Prove it fails* — run =sbcl --eval "(asdf:test-system :passepartout)"=
and confirm the new test fails (RED) before writing implementation
3. *Write the code* — modify the implementation in the same =.org= file
4. *Prove it passes* — run the test suite again, confirm GREEN
5. *Reflect* — ensure the test and code are both in the =.org= literate source
1. *Think in org* — write reasoning and goals in the .org file
2. *Write contract* — define each function's behavior in a ~** Contract~ section
3. *TDD from contract* — each contract item becomes a ~fiveam:test~; prove RED then GREEN
4. *Reflect in org* — ensure implementation is in .org source
5. *Update literate prose* — explain the code: what, why, how it connects
For *existing code* that lacks tests: write a characterization test that
captures current behavior as the spec. Then refactor.
* Literate Programming
No test may be committed without proof it was first run to failure.
~.org~ files in ~org/~ are the source of truth. ~lisp/~ files are generated by ~org-babel-tangle~.
* Literate Granularity
We strictly adhere to Literate Programming using Org-mode.
- *Never* edit `.lisp` files in `src/` directly.
- Modify the corresponding `.org` files in the `literate/` or `skills/` directories.
- Run `org-babel-tangle` to generate the source code.
- Every architectural decision, constraint, and implementation detail must be documented alongside the code in the `.org` file.
- Never edit =lisp/= files directly — always modify the corresponding =org/= file
- All ~#+begin_src lisp~ blocks in a file inherit their tangle destination from the file-level ~#+PROPERTY: header-args:lisp :tangle ../lisp/FILE.lisp~
- Every architectural decision, constraint, and implementation detail must be documented alongside the code
* Contracts and Tests
Every code change starts with a contract and a failing test. Write a ~** Contract~ section listing each function's behavior, then create a ~fiveam:test~ in the ~* Test Suite~ section for each contract item.
To run tests for a specific file:
#+begin_src bash
sbcl --noinform \
--eval '(load (merge-pathnames "quicklisp/setup.lisp" (user-homedir-pathname)))' \
--eval '(ql:quickload :passepartout :silent t)' \
--eval '(load "lisp/FILE.lisp")' \
--eval '(fiveam:run (intern "SUITE-NAME" :passepartout-TESTS))' --quit
#+end_src
No test may be committed without proof it was first run to failure (RED).
* Skill Creation Standard
Skills are the building blocks of Passepartout. They reside in the `skills/` directory.
A skill must define:
1. *Trigger*: A lambda determining if the skill should activate based on the context.
2. *Probabilistic Gate*: Optional. Generates a prompt for the LLM.
3. *Deterministic Gate*: A hardcoded Lisp function that guarantees safety or executes side-effects (the Dispatcher pattern).
A skill is a =.org= file in =org/= that defines:
Example Registration:
1. *Contract* — what the skill guarantees
2. *Implementation* — the code, tangled to ~lisp/~ via ~#+PROPERTY: header-args:lisp~
3. *Skill Registration* — a ~defskill~ form with ~:priority~, ~:trigger~, ~:probabilistic~ / ~:deterministic~
4. *Test Suite*~fiveam:test~ forms verifying the contract
Example:
#+begin_src lisp
(defskill :skill-example
(defskill :passepartout-example
:priority 100
:trigger (lambda (ctx) ...)
:probabilistic nil
:probabilistic (lambda (ctx) ...)
:deterministic (lambda (action ctx) ...))
#+end_src
* The Unified Envelope (Communication Protocol)
All inter-process communication occurs via the Unified Envelope. Do not use legacy specific types like `:CHAT`.
- Always use semantic types: `:REQUEST`, `:EVENT`, `:RESPONSE`, `:STATUS`, `:LOG`.
- Include routing metadata in the `:META` block (e.g., `(:SOURCE :TUI)`).
- Ensure generated `:REQUEST` messages include a mandatory `:TARGET` field.
* Project Structure
* Pull Request Process
1. Choose an Org file and write a failing test in its =* Test Suite= section.
2. Tangle and run to confirm RED (the test fails).
3. Write the implementation in the same Org file, tangle, run to confirm GREEN.
4. Ensure your working tree is clean.
5. Run the full test suite: =sbcl --eval "(asdf:test-system :passepartout)"=.
6. Submit a PR outlining the architectural intent and the specific Literate changes.
| Directory | Purpose |
|----------------------+--------------------------------------------------|
| =org/= | Literate source files (edit these) |
| =lisp/= | Tangled .lisp output (never edit) |
| =docs/= | ROADMAP, ARCHITECTURE, DESIGN_DECISIONS, etc. |
| =scripts/= | Build and utility scripts |
| ~/.local/share/passepartout/= | XDG data dir — deployed lisp files |
| ~/.config/passepartout/= | Config (.env) |
* Key Libraries
| Library | Purpose |
|------------------+----------------------------------|
| Croatoan | TUI (terminal UI) |
| usocket | TCP sockets (daemon protocol) |
| bordeaux-threads | Threading (reader thread) |
| dexador | HTTP client (LLM API calls) |
| cl-ppcre | Regex (search-files, dispatcher) |
| ironclad | SHA-256 (Merkle hashing) |
| hunchentoot | HTTP server |
| cl-json | JSON encoding/decoding |
* Protocol
All inter-process communication uses the Unified Envelope protocol over TCP (port 9105). Message types: ~:REQUEST~, ~:EVENT~, ~:RESPONSE~, ~:STATUS~, ~:LOG~. Each message includes a ~:META~ block with routing metadata.
* Pre-Commit Hook
Validates staged org files by tangling + structural-checking:
#+begin_src bash
ln -sf ../../scripts/pre-commit-repl-check .git/hooks/pre-commit
#+end_src
Runs automatically on ~git commit~.
* Testing Tools
** TUI REPL (~/eval~)
The TUI has a built-in command for live evaluation:
- ~/eval (+ 1 2)~ → result displayed in chat window
- ~/eval (add-msg :system "test")~ → inject a test message
** Tmux (TUI integration testing)
#+begin_src bash
tmux new-session -d -s test "passepartout tui 2>&1 | tee /tmp/tui.log"
tmux send-keys -t test "hello world" Enter
tmux capture-pane -t test -p -S -200
tmux kill-session -t test
#+end_src
** Swank (Emacs REPL for TUI)
1. Start TUI: ~passepartout tui~
2. In Emacs: ~M-x slime-connect RET 127.0.0.1 RET 4006~
3. ~C-M-x~ any form from =org/gateway-tui.org= → evaluates in live TUI process
4. Configure port: ~export TUI_SWANK_PORT=4009~ (default: 4006)