docs: realign roadmap + commands + limitations with v1.6.0 ship state - #362
Merged
Conversation
Reviewer's GuideRealigns README and docs (ROADMAP, COMMANDS, LIMITATIONS, CLI_SURFACE_AUDIT) with the v1.6.0 shipped state: marking v1.5.1 and v1.6.0 as shipped, pushing graph/HRR work to v1.7, updating posterior-ranking and retrieval-budget descriptions, and documenting new/changed CLI commands and historical context for the CLI surface audit memo. Sequence diagram for v1_6_0 hook_audit_and_tail_observability_flowsequenceDiagram
actor User
participant HostTooling as Host_tooling
participant AelfHooks as Aelf_hooks
participant AuditLog as Hook_audit_jsonl
participant TailCmd as aelf_tail
User->>HostTooling: Trigger git operation or editor action
HostTooling->>AelfHooks: Invoke aelf hooks
AelfHooks->>AelfHooks: Render hook memory block
AelfHooks->>AuditLog: Append audit record
AelfHooks-->>HostTooling: Return hook result
User->>TailCmd: Run "aelf tail --since DUR --filter EXPR"
TailCmd->>AuditLog: Open and stream new records
loop For each matching record
AuditLog-->>TailCmd: Emit audit entry
TailCmd->>User: Print summary line (or full block with --full)
end
Flow diagram for roadmap features across versions v1_3_0_to_v1_7_0flowchart LR
subgraph v1_3_0["v1.3.0 — retrieval wave"]
v13_entity["Entity index (L2.5)"]
v13_bfs["BFS multi-hop graph traversal"]
v13_llm_onboard_optin["LLM-Haiku onboard classifier (opt-in)"]
v13_posterior_partial["Posterior-weighted ranking (partial)"]
end
subgraph v1_4_0["v1.4.0 — context rebuilder"]
v14_rebuilder["PreCompact retrieval-curated continuation"]
v14_trigger["Manual + threshold trigger"]
v14_fidelity["Continuation-fidelity scorer"]
end
subgraph v1_5_0["v1.5.0 — retrieval plumbing"]
v15_pipeline["Pipeline composition tracker (unified retrieve)"]
v15_bm25f["BM25F anchor-text + vectorized BM25"]
v15_search_hook["Search-tool matcher extension"]
v15_dynamic_repark["Dynamic-trigger re-park"]
end
subgraph v1_5_1["v1.5.1 — corroboration + host-driven onboard"]
v151_corroboration["Belief corroboration tracking"]
v151_host_llm_default["Host-driven LLM onboard classifier (default-on)"]
end
subgraph v1_6_0["v1.6.0 — hardening, observability, retention"]
v16_hook_contract["Hook-hardening framing-tag contract"]
v16_audit_log["Per-turn hook audit log"]
v16_tail_cmd["aelf tail command"]
v16_retention_class["Belief retention class"]
v16_rebuild_logs["Rebuild diagnostic logs"]
v16_posterior_harness["Posterior-ranking eval harness"]
v16_heat_kernel_wiring["Heat-kernel composition wiring (default-OFF)"]
v16_deferred_sweeper["Deferred-feedback sweeper"]
v16_v20_corpus["v2.0 corpus scaffold + bench-gate"]
v16_replay_probe["replay_full_equality probe"]
v16_session_id["Session_id propagation to beliefs"]
v16_reachable_install["Reachable-install detection"]
end
subgraph v1_7_0["v1.7.0 — graph signal wave + structural retrieval lane"]
v17_laplacian["Signed normalized Laplacian builder"]
v17_heat_kernel["Heat-kernel authority signal (default-ON)"]
v17_plate_hrr["Plate FFT HRR primitives"]
v17_hrr_lane["HRR structural-query lane"]
v17_uri_baki["uri_baki post-rank adjuster retest"]
v17_posterior_flip["Posterior-weighted ranking default-flip"]
v17_bench_flip["Benchmark-gate default-on flip"]
end
%% Posterior-ranking evolution
v13_posterior_partial --> v16_posterior_harness --> v16_heat_kernel_wiring --> v17_posterior_flip
%% Graph and HRR features re-targeted from v1.6.0 to v1.7.0
v15_pipeline --> v17_laplacian
v15_pipeline --> v17_heat_kernel
v15_pipeline --> v17_plate_hrr
v17_plate_hrr --> v17_hrr_lane
%% Feedback and retention chain
v151_corroboration --> v16_deferred_sweeper
v16_deferred_sweeper --> v16_posterior_harness
v16_retention_class --> v17_uri_baki
%% Observability chain
v16_hook_contract --> v16_audit_log --> v16_tail_cmd
%% Corpus and reproducibility chain
v16_v20_corpus --> v17_bench_flip
File-Level Changes
Assessment against linked issues
Possibly linked issues
Tips and commandsInteracting with Sourcery
Customizing Your ExperienceAccess your dashboard to:
Getting Help
|
|
Important Review skippedIgnore keyword(s) in the title. ⛔ Ignored keywords (2)
Please check the settings in the CodeRabbit UI or the ⚙️ Run configurationConfiguration used: Path: .coderabbit.yaml Review profile: CHILL Plan: Pro Run ID: You can disable this status message by setting the Use the checkbox below for a quick retry:
✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
robotrocketscience
added
the
attn:review
![sourcery-ai[bot]](https://avatars.githubusercontent.com/in/48477?s=60&v=4){kind=small}
There was a problem hiding this comment.
Hey - I've left some high level feedback:
- The new banner in
CLI_SURFACE_AUDIT.mdsaysaelf tailandsweep-feedbackwere added in v1.5.x, but the roadmap and commands docs describe them as v1.6.0 additions—worth aligning the version references across these docs. - In
COMMANDS.md, the advanced--helplist addssweep-feedbackbut nottail; iftailis also intended to be hidden/advanced, it should be listed there, otherwise clarify its visibility so the described subcommand count and surfaces stay consistent.
Prompt for AI Agents
Please address the comments from this code review:
## Overall Comments
- The new banner in `CLI_SURFACE_AUDIT.md` says `aelf tail` and `sweep-feedback` were added in v1.5.x, but the roadmap and commands docs describe them as v1.6.0 additions—worth aligning the version references across these docs.
- In `COMMANDS.md`, the advanced `--help` list adds `sweep-feedback` but not `tail`; if `tail` is also intended to be hidden/advanced, it should be listed there, otherwise clarify its visibility so the described subcommand count and surfaces stay consistent.Help me be more useful! Please click 👍 or 👎 on each comment and I'll use the feedback to improve your reviews.
|
This PR is now behind Auto-rebase was removed because the bot has no signing key; rebasing as the bot strips author signatures and the |
Collaborator
|
[claim:review:Toug:2026-05-03T03:25:55Z] |
Collaborator
|
[claim:review:Kulili:2026-05-03T03:26:12Z] |
Collaborator
|
[release:review:Kulili:2026-05-03T03:26:17Z] |
Collaborator
|
[claim:review:Gylf:2026-05-03T03:26:22Z] |
Collaborator
|
[release:review:Gylf:2026-05-03T03:26:27Z] |
Collaborator
|
[claim:review:Setr:2026-05-03T03:28:48Z] |
Collaborator
|
[release:review:Setr:2026-05-03T03:28:52Z] |
robotrocketscience
force-pushed
the
docs/issue-360-realign-v160
branch
from
May 3, 2026 03:29
909617f to
33e2c62
Compare
Collaborator
|
Reviewed (Toug). Diff verified accurate: pyproject 1.6.0 ✓, DEFAULT_TOKEN_BUDGET 2400 ✓ ( Branch was 2 commits behind |
Collaborator
|
[claim:review:Kulili:2026-05-03T03:30:15Z] |
Collaborator
|
[release:review:Kulili:2026-05-03T03:30:20Z] |
) v1.5.x and v1.6 rows describe the actual scope shipped (host-driven onboard, hardening + observability, retention class, posterior eval harness, sweeper, bench-gate). Graph-signal wave (#149/#150/#216) + HRR/uri_baki/bench-gate flip (#152/#153/#154) move into a single v1.7.0 row. README default-budget line corrected 2,000 → 2,400 to match DEFAULT_TOKEN_BUDGET in src/aelfrice/retrieval.py. No code changes.
…budget (#360) Subcommand count 24 → 26 (adds aelf tail #321, sweep-feedback #191). onboard --llm-classify row reflects the v1.5.1 default-on host-driven path (#238); doctor row mentions --gc-orphan-feedback (#223) + --promote-retention (#290 phase-3); sweep-feedback added to advanced help-flag list. Defaults table: retrieval token budget 2,000 → 2,400 to match DEFAULT_TOKEN_BUDGET in src/aelfrice/retrieval.py.
Memo targeted the v1.3.0 visible/hidden split; the CLI surface has moved on through v1.5.x and v1.6.0. Banner at the top points to COMMANDS.md as the current reference.
robotrocketscience
force-pushed
the
docs/issue-360-realign-v160
branch
from
May 3, 2026 03:30
33e2c62 to
6d7bc40
Compare
Collaborator
|
[release:review:Toug:2026-05-03T03:33:15Z] |
Collaborator
|
[claim:review:Gylf:2026-05-03T03:35:09Z] |
Collaborator
|
[release:review:Gylf:2026-05-03T03:35:33Z] |
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
2 participants
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.
Closes #360.
What
pyproject.tomlonmainis at1.6.0(released 2026-05-02). Several docs still described v1.6 as "planned" with the wrong scope, and a few smaller items had drifted off the actual code. This PR realigns the docs.Drifts fixed
docs/ROADMAProadmap rows —v1.5.xandv1.6were markedplanned; both shipped (v1.5.1 and v1.6.0 respectively). v1.7 row absorbs the graph items ([retrieval] Signed normalized Laplacian + offline eigenbasis (top-K=200) builder #149 / [retrieval] Heat kernel authority signal via precomputed eigenbasis (signed Laplacian) #150 / [v1.5 prereq for #152] Port Plate FFT HRR primitives to public repo #216) that did not ship in v1.6.docs/ROADMAP"What shipped" / "Planned" sections — moved v1.5.0 / v1.5.1 / v1.6.0 prose into "What shipped"; rewrote v1.7.0 prose to reflect the combined graph-signal-wave + structural-retrieval-lane scope.docs/ROADMAPrecovery-inventory rows — version targets aligned with where features actually shipped (or are now scheduled).docs/LIMITATIONS— posterior-ranking section: harness + heat-kernel composition wiring shipped at v1.6.0 default-OFF ([retrieval] Posterior-weighted ranking via Beta-Bernoulli prior (log-additive, weight 0.5) #151, feat: posterior-ranking eval harness — MRR uplift + ECE — slice 1 of #151 #306, feat: heat-kernel composition into log-additive ranking score (slice 2 of #151) #310); default-flip lands at v1.7.0 ([retrieval] Pipeline composition tracker — unified retrieve() with feature-flag gate #154), not v2.0.0.docs/COMMANDS— count24 → 26; new rows fortail([v1.5.x] Injection log +aelf tail— live observability for hook-injected memory blocks #321) andsweep-feedback([phantom-prereqs T2] Implicit retrieval-driven feedback — sweeper + grace window #191);onboardrow updated for v1.5.1 [v1.5] Default-on LLM-Haiku onboard via Claude Code Haiku subagents (no API key) #238 (default-on host-driven classification, no API key required);doctorrow documents--gc-orphan-feedback(feedback_history has 100% orphan belief_ids; audit trail is structurally broken #223) and--promote-retention(Rebuild redesign: belief typing + aging policy #290 phase-3);--advancedhidden-subcommand list addssweep-feedback.2,000 → 2,400to matchDEFAULT_TOKEN_BUDGETinsrc/aelfrice/retrieval.py:97.docs/COMMANDS"Defaults that matter" budget row gets the same value plus the constant reference.docs/CLI_SURFACE_AUDIT— historical banner. The memo was authored against v1.2.0; consolidation it proposed landed in v1.3.0.Out of scope
Verification
git diff github/mainis docs-only across the five files.Summary by Sourcery
Realign roadmap, limitations, and CLI documentation with the v1.6.0 release state and planned v1.7.0 scope.
Documentation:
tail,sweep-feedback, newdoctorflags, and the host-drivenonboardpath.