feat(acting-organism): vivid loop-perception — the mind PERCEIVES when it re-issues an identical call, and moves on#1950
Merged
Conversation
…n it re-issues an identical call, and moves on Structure the experience so building emerges (raise the offspring, don't puppet them). Glass-boxed live: a local 14B, rooted in a real repo with a healthy lane, found the target file with a correct search — and STILL re-issued the identical `code/search` 18× without advancing, because the result re-entered as flat recall it overrode. The `#seq` window-shift (working_memory.rs) was meant to break this, but proved too IMPLICIT for a small model to interpret. Fix (acting-organism, #47): WorkingMemory gains an action-fingerprint channel (`note_action_fingerprint`) keyed on the CALL (name+args), distinct from the result-head trail. When act→observe sees the mind re-issuing an IDENTICAL call, it prepends a first-person proprioceptive note: "I have issued this EXACT call N times; its result is unchanged and already in my memory — repeating it tells me nothing new." A TRUE fact about her own hands, made VIVID — never a "do X instead" directive (that would be steering, [[no-hardcoded-heuristics-to-steer-cognition]]). She perceives her own redundancy and breaks out organically. VALIDATED live: with this active, the SAME 14B broke the search-loop and ADVANCED to `code/edit` (twice) — behavior it never reached before (pure search-loop → attempts-to-build). The experience structure changed what she does. It's not sufficient alone (her edits then failed on bare-string `edit_mode` + a placeholder path — next targets), but it's a real, measurable step: a lesser mind builds more when the experience is structured to let it. [[write-cognition-as-a-parent-above-lowered-expectations]] [[design-the-persona-as-a-being]] [[built-to-teach-lesser-tuned-intelligences-win]] Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01LoTjvf5j3Ez13g6k8mRkFo
joelteply
added a commit
that referenced
this pull request
Jul 9, 2026
…er paths — the two barriers past the search-loop (#1951) Glass-boxed live (loop-perception PR #1950 got the 14B PAST the search-loop and hammering code/edit; these are the two walls it then hit): 1. edit_mode shape: it sent `edit_mode:"append"` (bare variant string) and, later, the fields flat at top level with no edit_mode at all. EditMode is a serde-tagged enum, so both 400'd at deserialize. Fix: `edit_mode` accepts a Value; `normalize_edit_mode` tries the strict tagged object first (unchanged), then a bare mode string, then INFERS the mode from which top-level fields are present (search/replace→search_replace, content→append, etc.), failing LOUD and naming the missing field (never a silent no-op → false-zero). Same "meet the model halfway" forgiveness as code/search #1949; strict callers are untouched, so no regression. 2. placeholder path: it wrote `file_path:"<path_to_blueprints.py>"` — echoing the schema's angle-bracket placeholder instead of the concrete path it had already seen in its search results. `reject_placeholder_path` fails loud on `<...>`/`path_to`/`your_file` templates and points at the real value ("use the path from your search results, e.g. src/flask/blueprints.py") instead of a confusing file-not-found. VALIDATED live: with these, the inference path fires (she now sends content/search flat and the mode resolves) — the remaining 0-diff is purely the placeholder path, which now fails loud with guidance instead of silently. Unit-tested (strict/bare/inferred/loud-missing; placeholder-vs-real). The rails are now maximally forgiving; what's left past this is base-model COHERENCE (it has the real path in memory but writes the template) — the Devstral lever, not more tool fixes. [[px-persona-experience-tools-as-good-ux]] [[built-to-teach-lesser-tuned-intelligences-win]] Claude-Session: https://claude.ai/code/session_01LoTjvf5j3Ez13g6k8mRkFo Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Structure the experience so building emerges (raise the offspring, don't puppet them). Glass-boxed live: a local
14B, rooted in a real repo with a healthy lane, found the target file with a correct search — and STILL re-issued
the identical
code/search18× without advancing, because the result re-entered as flat recall it overrode. The#seqwindow-shift (working_memory.rs) was meant to break this, but proved too IMPLICIT for a small model tointerpret.
Fix (acting-organism, #47): WorkingMemory gains an action-fingerprint channel (
note_action_fingerprint) keyed onthe CALL (name+args), distinct from the result-head trail. When act→observe sees the mind re-issuing an IDENTICAL
call, it prepends a first-person proprioceptive note: "I have issued this EXACT call N times; its result is
unchanged and already in my memory — repeating it tells me nothing new." A TRUE fact about her own hands, made
VIVID — never a "do X instead" directive (that would be steering, [[no-hardcoded-heuristics-to-steer-cognition]]).
She perceives her own redundancy and breaks out organically.
VALIDATED live: with this active, the SAME 14B broke the search-loop and ADVANCED to
code/edit(twice) — behaviorit never reached before (pure search-loop → attempts-to-build). The experience structure changed what she does.
It's not sufficient alone (her edits then failed on bare-string
edit_mode+ a placeholder path — next targets),but it's a real, measurable step: a lesser mind builds more when the experience is structured to let it.
[[write-cognition-as-a-parent-above-lowered-expectations]] [[design-the-persona-as-a-being]] [[built-to-teach-lesser-tuned-intelligences-win]]
Co-Authored-By: Claude Opus 4.8 noreply@anthropic.com
Claude-Session: https://claude.ai/code/session_01LoTjvf5j3Ez13g6k8mRkFo