stabilize: snapshot-pinning scaffolding (Amara action #2 — completes Stabilize stage) by AceHack · Pull Request #223 · Lucent-Financial-Group/Zeta

AceHack · 2026-04-24T00:30:11Z

Summary

Completes Amara's Stabilize-stage from her 4th ferry roadmap (PR #221). The remaining action item: pin Claude model + loaded memory + prompt bundle so "Claude" isn't a moving target across model-version shifts.

What landed

Three-part v0 scaffolding:

tools/hygiene/capture-tick-snapshot.sh — capture helper (YAML / JSON output; CLI version + file SHAs + git state; agent fills model + prompt bundle hash)
docs/hygiene-history/session-snapshots.md — session-level + significant-event pin sidecar (append-only), seeded with first entry
docs/hygiene-history/loop-tick-history.md — new subsection documenting the snapshot-pinning discipline + relationship to the sidecar file

Amara Stabilize-stage complete

✓ Action Round 27 — plugin API + governance split + memory-in-repo #3 — decision-proxy-evidence schema (PR stabilize: decision-proxy-evidence schema + template (Amara action #3) #222)
✓ Action Round 28 — FsCheck LawRunner (Option B) + round-29 CI anchor #4 — branch-chat non-canonical framing (PR stabilize: decision-proxy-evidence schema + template (Amara action #3) #222)
✓ Action Round 26 — rename tail, §18 memory clarification, three dispatches #2 — snapshot pinning scaffolding (this PR)

First-run observation

Snapshot tool first fire surfaced: memory/MEMORY.md at 58,842 bytes (~58KB) — over the FACTORY-HYGIENE row #11 24,976-byte cap. Known separately-tracked drift. Not fixed in this PR; flagged for separate cleanup.

What this PR is NOT

Not CI-gated (v0 discipline; enforcement is Determinize-stage)
Not retroactive for prior sessions (starts from here forward)
Not complete — model_snapshot + prompt_bundle_hash are agent-filled / future-tool-filled respectively
Not a per-tick gate (per-tick pins are optional; don't slow CronList close)

Composes with "deterministic reconciliation" (Otto-67)

Snapshot pinning is the mechanism that reconciles "what Claude knew" with "what Claude did" across time. Necessary for audit, future tuning pipelines (per Otto-57 PR-review archive memory), and Amara-style drift analysis.

Test plan

capture-tick-snapshot.sh runs clean on current branch (YAML output seeded the sidecar entry)
Sidecar file + schema extension cross-reference each other and the capture tool
actionlint N/A (no new workflow); shellcheck-via-eyeball on the shell script

🤖 Generated with Claude Code

…(Amara action #2) Amara's 4th ferry (PR #221) action item #2: pin Claude model snapshot + loaded memory state + prompt bundle hash so "Claude" is not a moving target across model version shifts (3.5 → 3.7 → 4 → 4.x all have materially different system-prompt bundles + knowledge cutoffs + memory-retention language per archived Drive artifacts). Three-part scaffolding (v0): 1. tools/hygiene/capture-tick-snapshot.sh - Captures mechanically-accessible state: * Claude Code CLI version (`claude --version`) * CLAUDE.md + AGENTS.md + memory/MEMORY.md SHAs * Memory index byte count * Git HEAD + branch + repo * Date UTC * Per-user ~/.claude/CLAUDE.md SHA if present - Outputs YAML (default) or JSON (`--json`) - Agent fills model_snapshot + prompt_bundle_hash from session context (not CLI-accessible today) 2. docs/hygiene-history/session-snapshots.md - Session-level + significant-event pins (not per-tick) - Append-only row format: session_id / captured_utc / event (session-open | mid-session-pin | session-close | compaction) / agent / model / CLI version / git state / files SHAs / notes / prompt_bundle_hash - Seeded with one mid-session-pin for this tick's Otto-70 capture (the session has been running ~70 ticks; actual session-open is earlier and unreachable for pins) 3. docs/hygiene-history/loop-tick-history.md schema extension - New "On snapshot pinning" subsection documenting the relationship: per-tick pins optional + inline in `notes`; session-level pins go in the sidecar file. - Snapshot capture is discipline, not gate — don't slow the autonomous-loop tick-close for every fire. What the snapshot does NOT capture yet: - model_snapshot — known to the agent from session context, not exposed by `claude --version` (which gives CLI version only). Agent fills. - prompt_bundle_hash — no current tool reconstructs the system prompt bundle. Placeholder null until such a tool lands. Amara's Determinize-stage work potentially. - Active permissions / skill set — session-specific; not captured in v0. First run of the tool on this branch surfaced a separate drift: memory/MEMORY.md is at 58842 bytes (~58KB, over the FACTORY-HYGIENE row #11 24976-byte cap). Not fixed in this PR — known-separately tracked drift. Amara Stabilize-stage: 3/3 landed (with this PR). ✓ Action #3 — decision-proxy-evidence schema (PR #222) ✓ Action #4 — branch-chat non-canonical framing (PR #222) ✓ Action #2 — snapshot pinning scaffolding (this PR) "Deterministic reconciliation" framing (Otto-67 endorsement): snapshot pinning is the mechanism that reconciles "what Claude knew" with "what Claude did" across time — essential for any future audit, tuning, or Amara-style drift analysis. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

Copilot

Pull request overview

Adds initial snapshot-pinning scaffolding so autonomous-loop ticks and significant session events can record the exact “Claude state” (CLI version, key file SHAs, git state) to reduce model/prompt/memory drift over time.

Changes:

Introduces tools/hygiene/capture-tick-snapshot.sh to emit a snapshot fingerprint (YAML/JSON).
Adds docs/hygiene-history/session-snapshots.md as an append-only sidecar log for session-level state pins.
Extends docs/hygiene-history/loop-tick-history.md with guidance on when/how to pin snapshots.

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 5 comments.

File	Description
tools/hygiene/capture-tick-snapshot.sh	New capture helper that prints a snapshot fingerprint of git + key file SHAs + CLI version (YAML/JSON).
docs/hygiene-history/session-snapshots.md	New append-only session snapshot sidecar with schema and seeded first entry.
docs/hygiene-history/loop-tick-history.md	Documents optional per-tick snapshot pinning and links to the new sidecar + capture helper.

Copilot · 2026-04-24T00:33:12Z

+    git hash-object "$1" 2>/dev/null || printf ''
+  fi
+}
+
+safe_bytes() {
+  if [[ -f "$1" ]]; then
+    wc -c < "$1" 2>/dev/null | tr -d ' ' || printf ''
+  fi


P0: safe_sha/safe_bytes can return a non-zero status when the file does not exist (the [[ -f ... ]] test fails), which can still abort the script under set -e during var=$(safe_sha ...) assignments. Make these helpers always succeed (exit 0) and print an empty string when the file is missing, matching the comment’s stated behavior.

Suggested change

git hash-object "$1" 2>/dev/null || printf ''

fi

}

safe_bytes() {

if [[ -f "$1" ]]; then

wc -c < "$1" 2>/dev/null | tr -d ' ' || printf ''

fi

git hash-object "$1" 2>/dev/null || printf ''

else

printf ''

fi

return 0

}

safe_bytes() {

if [[ -f "$1" ]]; then

wc -c < "$1" 2>/dev/null | tr -d ' ' || printf ''

else

printf ''

fi

return 0

Copilot · 2026-04-24T00:33:13Z

+memory_index_bytes=$(safe_bytes "memory/MEMORY.md")
+head_sha=$(git rev-parse HEAD 2>/dev/null || printf 'unknown')
+branch=$(git rev-parse --abbrev-ref HEAD 2>/dev/null || printf 'unknown')
+repo_full=$(git config --get remote.origin.url 2>/dev/null | sed -E 's|.*[:/]([^/]+/[^/]+)(\.git)?$|\1|' || printf 'unknown')


P1: repo_full parsing can incorrectly retain the .git suffix for common remotes like git@github.com:owner/repo.git or https://.../owner/repo.git because the capture group includes the .git portion. Consider stripping a trailing .git after extraction (or use git remote get-url origin + explicit suffix removal) so repo is consistently owner/repo.

Suggested change

repo_full=$(git config --get remote.origin.url 2>/dev/null | sed -E 's|.*[:/]([^/]+/[^/]+)(\.git)?$|\1|' || printf 'unknown')

repo_full=$(git config --get remote.origin.url 2>/dev/null | sed -E 's|.*[:/]([^/]+/[^/]+)(\.git)?$|\1|' | sed -E 's|\.git$||' || printf 'unknown')

Copilot · 2026-04-24T00:33:13Z

+# Captures a snapshot pin of factory state at tick-open /
+# tick-close time. Prints a YAML fragment that can be pasted
+# into:
+#   - docs/hygiene-history/session-snapshots.md (session-level)
+#   - docs/decision-proxy-evidence/DP-NNN.yaml `model` block
+#     (decision-level)
+#   - a tick-history row's `notes` column (tick-level)


P1: The header claims this YAML can be pasted into session-snapshots.md and the DP model block, but the script’s YAML output is a snapshot: object (and JSON has top-level model_snapshot), which doesn’t match either documented schema (session snapshots are list rows; DP expects a nested model: block). Align the emitted shape with the target schemas (or add mode flags) to avoid copy/paste errors.

Copilot · 2026-04-24T00:33:13Z

+per-decision pin-time. Addresses Amara's 4th-ferry concern
+(PR #221) that "Claude is not a single stable operator unless
+the actual snapshot, system-prompt bundle, and loaded memory
+surfaces are all pinned and recorded."
+
+## Why this file
+
+Across Claude model versions (3.5 → 3.7 → 4 → 4.x), the
+system-prompt bundle + knowledge cutoff + memory-retention
+language shift materially. When a future session, external
+reviewer, or tuning pipeline asks *"what did Kenji actually
+know when this decision was made?"* this file answers.


P1: This doc introduces direct contributor/agent names (e.g., “Kenji”, “Amara”, “Otto”). docs/AGENT-BEST-PRACTICES.md operational standing rule forbids name attribution in non-exempt docs and requires role references instead. Please rewrite these references to role-refs (e.g., “architect”, “external AI maintainer”, “loop agent”, “human maintainer”) and keep personal names confined to the exempt surfaces.

Suggested change

per-decision pin-time. Addresses Amara's 4th-ferry concern

(PR #221) that "Claude is not a single stable operator unless

the actual snapshot, system-prompt bundle, and loaded memory

surfaces are all pinned and recorded."

## Why this file

Across Claude model versions (3.5 → 3.7 → 4 → 4.x), the

system-prompt bundle + knowledge cutoff + memory-retention

language shift materially. When a future session, external

reviewer, or tuning pipeline asks *"what did Kenji actually

know when this decision was made?"* this file answers.

per-decision pin-time. Addresses the external AI maintainer's

4th-ferry concern (PR #221) that "Claude is not a single

stable operator unless the actual snapshot, system-prompt

bundle, and loaded memory surfaces are all pinned and

recorded."

## Why this file

Across Claude model versions (3.5 → 3.7 → 4 → 4.x), the

system-prompt bundle + knowledge cutoff + memory-retention

language shift materially. When a future session, external

reviewer, or tuning pipeline asks *"what did the architect

actually know when this decision was made?"* this file

answers.

Copilot · 2026-04-24T00:33:14Z

+scaffolding (PR <this-pr>): when a tick's action is
+proxy-significant or settings-changing, the `notes`
+column can include a brief snapshot fingerprint
+(CLAUDE.md SHA, model snapshot). For session-level
+state — model swap, compaction boundary, significant
+memory migration — use the dedicated sidecar file


P1: PR <this-pr> is a placeholder that will become stale/unresolvable after merge. Replace it with the actual PR number (and ideally a link) so the provenance trail remains auditable.

Suggested change

scaffolding (PR <this-pr>): when a tick's action is

proxy-significant or settings-changing, the `notes`

column can include a brief snapshot fingerprint

(CLAUDE.md SHA, model snapshot). For session-level

state — model swap, compaction boundary, significant

memory migration — use the dedicated sidecar file

scaffolding: when a tick's action is proxy-significant

or settings-changing, the `notes` column can include a

brief snapshot fingerprint (CLAUDE.md SHA, model

snapshot). For session-level state — model swap,

compaction boundary, significant memory migration — use

the dedicated sidecar file

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 040b849a8e

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

chatgpt-codex-connector · 2026-04-24T00:34:03Z

+claude_md_sha=$(safe_sha "./CLAUDE.md")
+agents_md_sha=$(safe_sha "./AGENTS.md")
+memory_index_sha=$(safe_sha "memory/MEMORY.md")
+memory_index_bytes=$(safe_bytes "memory/MEMORY.md")


Anchor snapshot file paths to repository root

The snapshot fields are read from ./CLAUDE.md, ./AGENTS.md, and memory/MEMORY.md relative to the caller’s current directory, so running this helper from any non-root cwd (for example docs/) silently produces empty hashes/byte counts while still returning success. That creates incomplete pin records that appear valid and weakens the audit/reproducibility goal of this feature; resolve paths from git rev-parse --show-toplevel (or the script directory) before hashing.

Useful? React with 👍 / 👎.

…rows (Amara Govern-stage 1/2) Amara's 4th ferry (PR #221 absorb) named populating docs/CONTRIBUTOR-CONFLICTS.md as the Govern-stage action: the schema has existed since PR #166 but the Resolved table was empty despite multiple session-observed contributor-level disagreements that closed with evidence. Backfills three genuine contributor-level conflicts observed this session (narrow scope — not maintainer-directives, which are out-of-scope per the schema's contributor-level disagreement definition): - CC-001: Copilot (PR reviewer) vs Aaron on no-name-attribution rule scope (history-file exemption). Resolved in Aaron's favor via Otto-52 clarification; policy BACKLOG row filed in PR #210. - CC-002: Amara (4th ferry) vs Otto (pre-Otto-67 pattern) on Stabilize-vs-keep-opening-new-frames. Resolved in Amara's favor; Otto pivoted at Otto-68 to execute her roadmap; 3/3 Stabilize + 3/5 Determinize landed via PRs #222/#223/#224/#225/#226. - CC-003: Codex (PR reviewer) vs Otto (initial framing) on citing-absent-artifacts. Resolved in Codex's favor via fix commits 29872af/1c7f97d on PRs #207/#208; pattern now discipline (distinguish merged-on-main from proposed-in-PR-open). All three rows follow the schema's 8-column layout and include the full Resolution-so-far / Scope / Source cells the schema requires. No retroactive Aaron→human-maintainer sweep of prior rows; schema's rule 1 (resolutions are additive) honored. This is 1/2 of Amara's Govern-stage work. 2/2 is the authority-envelope + escalation-path ADR (deferred, M-effort). Part of Amara's 4-stage remediation roadmap (Stabilize → Determinize → Govern → Assure). Otto-75 tick.

…rows (Amara Govern 1/2) (#227) * govern: CONTRIBUTOR-CONFLICTS backfill — 3 resolved session-observed rows (Amara Govern-stage 1/2) Amara's 4th ferry (PR #221 absorb) named populating docs/CONTRIBUTOR-CONFLICTS.md as the Govern-stage action: the schema has existed since PR #166 but the Resolved table was empty despite multiple session-observed contributor-level disagreements that closed with evidence. Backfills three genuine contributor-level conflicts observed this session (narrow scope — not maintainer-directives, which are out-of-scope per the schema's contributor-level disagreement definition): - CC-001: Copilot (PR reviewer) vs Aaron on no-name-attribution rule scope (history-file exemption). Resolved in Aaron's favor via Otto-52 clarification; policy BACKLOG row filed in PR #210. - CC-002: Amara (4th ferry) vs Otto (pre-Otto-67 pattern) on Stabilize-vs-keep-opening-new-frames. Resolved in Amara's favor; Otto pivoted at Otto-68 to execute her roadmap; 3/3 Stabilize + 3/5 Determinize landed via PRs #222/#223/#224/#225/#226. - CC-003: Codex (PR reviewer) vs Otto (initial framing) on citing-absent-artifacts. Resolved in Codex's favor via fix commits 29872af/1c7f97d on PRs #207/#208; pattern now discipline (distinguish merged-on-main from proposed-in-PR-open). All three rows follow the schema's 8-column layout and include the full Resolution-so-far / Scope / Source cells the schema requires. No retroactive Aaron→human-maintainer sweep of prior rows; schema's rule 1 (resolutions are additive) honored. This is 1/2 of Amara's Govern-stage work. 2/2 is the authority-envelope + escalation-path ADR (deferred, M-effort). Part of Amara's 4-stage remediation roadmap (Stabilize → Determinize → Govern → Assure). Otto-75 tick. * govern: annotate CC-002/CC-003 Source cells — PR #221/#219 open, not yet on main Applies CC-003's own discipline (cite-as-open-not-landed) to CC-002 and CC-003 themselves. Both rows cited `docs/aurora/2026-04-23-amara-memory-drift-*` and `docs/aurora/2026-04-23-amara-decision-proxy-*` without the "not yet on main" marker — the files are added by PRs #221 / #219 which are still open. Drain for PR #227 review threads PRRT_kwDOSF9kNM59RFIx and PRRT_kwDOSF9kNM59RFJE (dangling file refs at lines 132, 133). * fix: markdownlint auto-fixes on governance doc Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.7 <noreply@anthropic.com>

Copilot AI review requested due to automatic review settings April 24, 2026 00:30

AceHack enabled auto-merge (squash) April 24, 2026 00:30

Copilot started reviewing on behalf of AceHack April 24, 2026 00:30 View session

AceHack merged commit a1a013f into main Apr 24, 2026
12 checks passed

AceHack deleted the stabilize/snapshot-pinning-tick-history-amara-action-2 branch April 24, 2026 00:31

Copilot AI reviewed Apr 24, 2026

View reviewed changes

chatgpt-codex-connector Bot reviewed Apr 24, 2026

View reviewed changes

AceHack mentioned this pull request Apr 24, 2026

govern: CONTRIBUTOR-CONFLICTS backfill — 3 resolved session-observed rows (Amara Govern 1/2) #227

Merged

4 tasks

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

stabilize: snapshot-pinning scaffolding (Amara action #2 — completes Stabilize stage)#223

stabilize: snapshot-pinning scaffolding (Amara action #2 — completes Stabilize stage)#223
AceHack merged 1 commit intomainfrom
stabilize/snapshot-pinning-tick-history-amara-action-2

AceHack commented Apr 24, 2026

Uh oh!

Uh oh!

Copilot AI left a comment

Uh oh!

Copilot AI Apr 24, 2026

Uh oh!

Copilot AI Apr 24, 2026

Uh oh!

Copilot AI Apr 24, 2026

Uh oh!

Copilot AI Apr 24, 2026

Uh oh!

Copilot AI Apr 24, 2026

Uh oh!

chatgpt-codex-connector Bot left a comment

Uh oh!

chatgpt-codex-connector Bot Apr 24, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

	repo_full=$(git config --get remote.origin.url 2>/dev/null \| sed -E 's\|.*[:/]([^/]+/[^/]+)(\.git)?$\|\1\|' \|\| printf 'unknown')
	repo_full=$(git config --get remote.origin.url 2>/dev/null \| sed -E 's\|.*[:/]([^/]+/[^/]+)(\.git)?$\|\1\|' \| sed -E 's\|\.git$\|\|' \|\| printf 'unknown')

Conversation

AceHack commented Apr 24, 2026

Summary

What landed

Amara Stabilize-stage complete

First-run observation

What this PR is NOT

Composes with "deterministic reconciliation" (Otto-67)

Test plan

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Copilot AI Apr 24, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 24, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 24, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 24, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 24, 2026

Choose a reason for hiding this comment

Uh oh!

chatgpt-codex-connector Bot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

chatgpt-codex-connector Bot Apr 24, 2026

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants