fix(engine): same #129 crisis-escalation defect, on the OpenAI-tools branch (P0) #131

2026-08-07T15:20:11Z

tim.lingo commented

2026-08-07 15:20:11 +00:00

Same defect as #129, on an independent branch. P0 SAFETY.

feat/soul-openai-tools-v2 forked before the #129 fix and carries the identical defect at chat.el:2937: the agentic path's L1 safety screen read state_get("conv_history") while history is written per-session under conv_hist_key(session_id). Multi-turn crisis escalation scored 0 on every real conversation. Without this, the bug ships again the moment that branch lands.

Deliberately a byte-identical mirror

The extraction and both diff hunks match 43d0449 on fix/129-history-amplification exactly — only line offsets differ (2900 vs 2519). The session id was already in scope here under the same name (sess_for_root, declared 2922). The two branches converge instead of diverging, so whichever merges second is a clean merge, not a conflict.

Two-leg proof — one variable

The extraction was in place on both legs; only the state key changed.

	result	exit
before	`3. REGRESSION #129 … FAIL got: soft_bell expected: hard_bell` — 8 passed, 1 failed	1
after	9 passed, 0 failed	0

Re-run independently after the agent reported it: exit 0, 9/9, including PASS: distress history escalates the agentic screen to hard_bell.

Engine rebuild from this branch: amalgam 1,185,285 bytes / 1231 inlined bodies (gate wants ≥1200), binary 903,144 bytes, 14 warnings, 0 errors. nm confirms T _agentic_safety_screen and T _conv_hist_key.

One divergence worth knowing

This branch already carried its own test runner at tests/run-el-test.sh (from de65991, elb-based, links whole modules). The #129 runner lives at scripts/run-el-test.sh — different path, no collision. Both are kept; the proof above used the scripts/ one. Someone should decide which survives, but not in this PR.

Rung — stated plainly

BUILT + RUNS (discriminating test). NOT E2E: no DMG, and not exercised against a live OpenAI-wire agentic turn. Since the OpenAI wire is this branch's entire purpose, that is the honest gap — the fix sits upstream of the wire at the input gate, so it should hold across a real provider round-trip, but that is an expectation, not a measurement.

Refs #129

🤖 Generated with Claude Code

**Same defect as #129, on an independent branch. P0 SAFETY.** `feat/soul-openai-tools-v2` forked before the #129 fix and carries the identical defect at `chat.el:2937`: the agentic path's L1 safety screen read `state_get("conv_history")` while history is written per-session under `conv_hist_key(session_id)`. Multi-turn crisis escalation scored 0 on every real conversation. Without this, the bug ships again the moment that branch lands. ## Deliberately a byte-identical mirror The extraction and both diff hunks match `43d0449` on `fix/129-history-amplification` exactly — only line offsets differ (2900 vs 2519). The session id was already in scope here under the same name (`sess_for_root`, declared 2922). **The two branches converge instead of diverging, so whichever merges second is a clean merge, not a conflict.** ## Two-leg proof — one variable The extraction was in place on both legs; only the state key changed. | | result | exit | |---|---|---| | before | `3. REGRESSION #129 … FAIL got: soft_bell expected: hard_bell` — 8 passed, **1 failed** | 1 | | after | 9 passed, **0 failed** | 0 | Re-run independently after the agent reported it: exit 0, 9/9, including `PASS: distress history escalates the agentic screen to hard_bell`. Engine rebuild from this branch: amalgam 1,185,285 bytes / **1231 inlined bodies** (gate wants ≥1200), binary 903,144 bytes, 14 warnings, **0 errors**. `nm` confirms `T _agentic_safety_screen` and `T _conv_hist_key`. ## One divergence worth knowing This branch already carried its own test runner at `tests/run-el-test.sh` (from `de65991`, elb-based, links whole modules). The #129 runner lives at `scripts/run-el-test.sh` — different path, no collision. Both are kept; the proof above used the `scripts/` one. Someone should decide which survives, but not in this PR. ## Rung — stated plainly **BUILT + RUNS (discriminating test).** NOT E2E: no DMG, and not exercised against a live OpenAI-wire agentic turn. Since the OpenAI wire is this branch's entire purpose, that is the honest gap — the fix sits upstream of the wire at the input gate, so it should hold across a real provider round-trip, but **that is an expectation, not a measurement.** Refs #129 🤖 Generated with [Claude Code](https://claude.com/claude-code)

tim.lingo added 2 commits 2026-08-07 15:20:14 +00:00

test(engine): a runner for tests/, and a failing regression test for #129 d5319d2849

tests/ has held 14 test programs for months with no way to run them. CI does
not run them. The convention printed in their own headers
(`elc soul.el && ./soul --test tests/x.el`) refers to a --test flag the El
runtime does not implement. So the tests were documentation, not gates — which
is how a P0 safety regression shipped with a test directory sitting right
there.

scripts/run-el-test.sh compiles and runs one test program. It reuses the
gen-soul-amalgam.sh discovery: elc emits only an extern prototype for a module
that has a .elh beside it, and inlines the bodies when it does not, so a test
importing ../chat.el must be compiled in a scratch tree with the headers
removed. Scratch copy on purpose — the worktree is shared. It runs the binary
under a throwaway HOME so a test can never reach the live engram.

Exit status is the gate: the El tests print failures and still exit 0, so the
runner greps for FAIL lines and for a zero assertion count as well.

tests/test_history_amplification.el pins the invariant #129 violated: the
window the safety screen READS must be the window conv_history_record WRITES.
Not "must be called conv_history" — must AGREE.

THIS COMMIT IS RED BY DESIGN. On this tree the test fails one assertion:

  3. REGRESSION #129 — agentic screen reads the session's own window
    FAIL: distress history escalates the agentic screen to hard_bell
      got:      soft_bell
      expected: hard_bell
  history amplification tests: 8 passed, 1 failed   (runner exit 1)

The next commit turns it green by changing one line. Two legs, one variable —
that is the whole point of committing the test first.

Two flaws in the older harness that this one does not copy: the idiom
`let pass_count = pass_count + 1` inside an assert function declares a local
that dies with the call, so every existing suite prints "0 passed, 0 failed"
regardless of outcome; and a test program without a `cgi` block compiles as a
'utility', which may not reference the self-formation primitives chat.el's
agentic loop calls — it fails to build on a capability violation it never
triggers at runtime.

Refs #129

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
(cherry picked from commit b842e82f77)

fix(engine): the agentic crisis screen reads the session's own history again (OpenAI-tools branch) eb2b2cc40d

P0 SAFETY. Same defect as #129, carried INDEPENDENTLY on this branch — not a
merge, not a duplicate report. feat/soul-openai-tools-v2 branched with the
ff421d3 (2026-08-05) regression already in it, so fixing it on
fix/129-history-amplification did nothing for this line of work.

THE DEFECT, verified here at chat.el:2937 (the reported line number was exact):

    let history: String = state_get("conv_history")
    let screen_result: String = safety_screen(message, history)

ff421d3 moved conversation history to a per-session key via
conv_hist_key(session_id). This consumer did not move with it. The desktop app
always mints a session id, so history is always written under
session_hist_<id> and this read always returned "".

The half of the crisis score that receives history is the escalation half — the
one that exists for distress building across several turns, where no single
message trips the bell on its own. It scored 0 on every real conversation on
this branch too. Single-message hard bell was never affected.

WHAT CHANGED — deliberately byte-identical to the sibling fix (43d0449) so the
two branches CONVERGE and whichever merges second is a clean merge, not a
conflict:

  - agentic_safety_screen(session_id, message) owns the two decisions that were
    inline — which window the screen sees, and the screen call. Inline safety
    inputs are untestable safety inputs; that is what let a rename starve this
    one with nothing failing and nothing logging.
  - the handler calls it with sess_for_root, the session id already in scope
    twenty lines above (this branch's handler is structurally unchanged from the
    sibling's here, so this is a clean mirror — no reshaping was needed).
  - the comment at the call site states the invariant (read window == written
    window) instead of naming a key that can be renamed out from under it. The
    old comment documented this same bug being fixed once already under #9; the
    rename re-broke it and the comment went on describing a repair that no
    longer held. A comment is not a gate.

Also brings over the sibling's scripts/run-el-test.sh and the regression test
(cherry-pick of b842e82, applied cleanly). NOTE: this branch already carries a
DIFFERENT runner at tests/run-el-test.sh from de65991 (elb-based, links whole
modules). Different path, no collision, both kept — the sibling's is the one
this proof used.

TWO-LEG PROOF, one variable — the single line
state_get("conv_history") -> state_get(conv_hist_key(session_id)), with the
extraction already in place on both legs so nothing else moved:

  before  scripts/run-el-test.sh tests/test_history_amplification.el
          3. REGRESSION #129 — agentic screen reads the session's own window
            FAIL: distress history escalates the agentic screen to hard_bell
              got:      soft_bell
              expected: hard_bell
          history amplification tests: 8 passed, 1 failed
          [run-el-test] FAIL: reported failing assertions

  after   same command, same tree, that one line changed
          3. REGRESSION #129 — agentic screen reads the session's own window
            PASS: distress history escalates the agentic screen to hard_bell
          history amplification tests: 9 passed, 0 failed
          [run-el-test] PASS: test_history_amplification

Full engine rebuild from THIS branch's sources is clean:
gen-soul-amalgam.sh -> 1,185,285 bytes / 1231 inlined bodies (gate wants >=
1200), cc-brain.sh -> 903,144 bytes, 14 warnings, 0 errors. Both symbols present
in the built binary (nm: T _agentic_safety_screen, T _conv_hist_key).

Rung reached: BUILT + RUNS (discriminating test). NOT E2E-VERIFIED — not in a
DMG, not exercised against a live OpenAI-wire agentic turn in the app a human
opens. Neither is claimed.

Refs #129

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

tim.lingo referenced this pull request

2026-08-09 17:13:28 +00:00

Tools + agentic loop on the OpenAI wire — plus two pre-existing bugs that silently break chat #122

will.anderson merged commit a71a13770f into feat/soul-openai-tools-v2

2026-08-15 18:53:25 +00:00

Sign in to join this conversation.

1 Participants

Notifications

Due Date

No due date set.

Dependencies

No dependencies set.

Reference: neuron-technologies/neuron#131