Skip to content

Releases

v0.7.1

Choose a tag to compare

@github-actions github-actions released this 19 Oct 10:04

Fixed

  • Test fixture neutralized: the double-dash Windows path sample no longer mirrors a real
    drive layout (no functional change).

v0.7.0

Choose a tag to compare

@github-actions github-actions released this 19 Oct 09:13

Added

  • Concept tables from a file: conceptsFile loads a runtime-generated table
    (concepts / deprecated / filtered / scope / watermark) as read-only user data —
    ~-expanded, mtime+size cached, capped at 1 MB and bounded field by field; inline
    concepts win on conflicts. mode: "concepts" reports the source, scope, watermark,
    keyword count and filtered-word count.
  • /memory elevation command: prints a scoped work package for the keyword-table workflow in
    docs/ELEVATION.md — row count, window count, the 2,000-row budget brake, the vocabulary
    watermark and how many rows are newer than it. Numbers only: no block text is pulled into the
    session by the command itself.
  • Bounded timeline preview: preview: true appends one redacted sentence (≤120 chars) per
    line and is honoured only for windows of at most 80 entries.
  • docs/ELEVATION.md: the keyword-table workflow (scope → deconstruct → cluster/keywordize →
    screen → outline/tasks → incremental), keyword rules K1–K6, and the elevation artifact schema.
    Filtering is a per-scope judgment recorded in the table's filtered list — the repository
    hardcodes no stopword list. This document is repository-only and is not shipped in the package.

Changed

  • sanitizeCfg accepts conceptsFile (≤500 chars) and merges the table with inline vocabulary.
  • Related-evidence stopwords moved to the stopwords config key (user data; empty by default).
    The bundle now ships no word list at all.
  • The memory_search prompt guidance is slimmed to calling-time rules; the keyword-table workflow
    lives in docs/ELEVATION.md and the /memory elevation command instead of the tool prompt.

Removed

  • memory_get tool removed (breaking change): block + detail: true on memory_search
    replaces it — registration, schema and prompt references removed.

Fixed

  • Vocabulary hints now append to timeline / related / decompose results too (previously
    only the default and concept paths) and consider both concept and query at once.
  • A broken conceptsFile is now cached after the first failure (mtime+size, or a stable entry
    when the file disappears) instead of being re-read and re-logged on every search.
  • mode: "concepts" now reports config / file / config+file based on what the file
    actually contributes, not just on whether inline concepts exist.

v0.6.2

Choose a tag to compare

@github-actions github-actions released this 19 Oct 09:13

Added

  • Runtime vocabulary guidance: deprecated in pi-billion-memory.json maps a retired term to
    its replacement (empty string = none). When a query contains one, memory_search appends a
    bounded Vocabulary hints line, and mode: "concepts" lists the table beside the configured
    concepts. The prompt guidance now tells the agent to record concept mappings and retired terms
    there, so the vocabulary is generated at runtime instead of being hardcoded.

Changed

  • sanitizeCfg bounds the deprecated table (MAX_DEPRECATED_TERMS, per-term length caps) and
    drops non-string replacements.

v0.6.1

Choose a tag to compare

@github-actions github-actions released this 18 Oct 08:54
78b8e2c

Added

  • Concept search (user-defined): memory_search takes a concept name that
    expands to the concrete terms the corpus actually uses — the vocabulary comes
    from the concepts key of pi-billion-memory.json (the extension ships none),
    searches every term, merges the hits, ranks by how many distinct terms matched,
    and reports Matched terms; mode: "concepts" lists what is configured.
  • Requirement decomposition: decompose: true splits a conjunctive query
    (Chinese commas/semicolons, and) into clauses and searches each clause
    separately, grouping the hits per clause.
  • timeline accepts from/to as ISO 8601 strings (date-only = the local day) as
    well as epoch milliseconds; an unparsable bound is rejected instead of ignored.

Changed

  • timeline renders local time (the header states times local UTC±hh:mm), tags
    non-pi entries with [opencode]/[bili], and reports listed/total with merged
    duplicates; limit: 0 returns the full catalog and a positive limit keeps the
    newest N instead of silently slicing.

Removed

  • The built-in concept vocabulary (src/concepts.data.ts). Vocabulary is user
    runtime data, not repository content (AGENTS.md §1), so 0.6.1 retires the
    bundled table: define concepts under "concepts" in pi-billion-memory.json.

v0.6.0

Choose a tag to compare

@github-actions github-actions released this 17 Oct 20:25

Added

  • bili-session adapter: indexes billion-context proxy session files
    (<host>_<hash>.json, blocks under payload.state.blocks, e.g. WorkBuddy/codebuddy sessions).
    The project name is the file's <host> segment, or the envelope's own upstreamOrigin/protocol
    when present; bili blocks are stored as not expandable, like opencode state files.
  • Upstream lifecycle/lineage metadata is now stored when the source provides it: active,
    direct_block_ids (compression lineage), duration_ms, generation, survived_count, and
    compress_call_id; sources also records format_version/producer. A force rescan refreshes
    these fields in place without touching stored summaries.
  • memory_search gains a mode: timeline returns the complete memory catalog (one line per
    memory — time, project, block id, tier, topic — date-grouped, never truncated, unplaced rows
    included), and related compares the sentence containing the query words in each block (with
    conservative stopwords and a data-driven IDF gate against generic-token noise) and shows that
    evidence sentence. The default blocks behavior is unchanged. A new read-only memory_get tool
    opens a catalog/search line and returns the stored, already-redacted summary (never raw messages).
  • Optional expansion (expandEnabled) now also works for billion-context sessions: memory_expand
    reads the envelope's own payload.blockContents (full/one manifest) instead of the host session
    log. It stays two-step, bounded, redacted and read-only.
  • Identical memories copied across source files (fork/copy artifacts) merge into a single result
    that lists the alternate sources; blocks from the same source are never collapsed, so multi-tier
    parent/child pairs survive.
  • related compares a bounded text window (minimum 20, maximum 800 characters) at sentence level,
    drops candidates with no query-word sentence or no evidence, and never deletes content words;
    generic tokens are demoted by the IDF gate. A new excludePaths config skips test/temp source
    paths across every adapter.

Changed

  • CONTRIBUTING.md documents the main branch-protection rule and how to lift force-push
    protection for the length of a deliberate history rewrite.
  • Upgraded the pi compatibility test target from @earendil-works/pi-coding-agent 0.85.1 to
    1.3.0; the extension continues to use only public extension APIs. This also picks up undici
    8.10.2 and its security fixes.
  • Adapters now honor upstream format versions: billion-context-pi sidecars carry schemaVersion
    (missing = v1) and billion-context session envelopes carry version (1-3). An unknown newer
    version is skipped and logged once, never rewritten, and never advances the watermark.
  • Auxiliary store files that live next to session files (.content-store.json, *.fb.json,
    SQLite files) are skipped without being read, so a broad allow-list glob cannot turn them into
    repeating failed sources.
  • Block expansion now reads the session only until the referenced messages are found and falls
    back to a tail window when the references sit near the end of a long file; the byte budget is
    unchanged. pi-sidecar allow-list entries also accept optional project/cwd hints that skip
    the lazy session-header read.
  • Model-facing output (memory_search results, related evidence, memory_get, and the timeline
    topic) is re-passed through the secret filter as defense-in-depth, so a row ingested before the
    current filter cannot surface a value that would be redacted today. Stored rows are not rewritten
    (delete *.db* and rescan to scrub durably).
  • Hardening for untrusted input: upstream metadata (project label, producer,
    compress_call_id, lineage ids) is length-capped and generation accepts only young/old;
    the query is length/token capped; the full timeline catalog refuses (with a hint) above 20000
    rows instead of silently truncating or exhausting memory.

v0.5.2

Choose a tag to compare

@github-actions github-actions released this 10 Sep 13:06

Fixed

  • A background scan can no longer reopen the store after session_shutdown, including a shutdown
    that lands while the allow-list is being read: the scan captures the session generation before
    its first await, and MemoryDb.open() refuses a latched store (getDb() already did).
  • Allow-list patterns are matched against paths relative to the source root, so a directory prefix
    (sub/*.json, 2026/**/*.jsonl.acp.json) narrows the scan instead of silently widening it into
    a recursive scan of the whole root.
  • Symlinks are ignored consistently. The non-recursive listing now uses withFileTypes like the
    recursive walk, so a link inside the root can no longer pull in a file from outside the
    allow-list.
  • An unreadable or missing allow-list directory is counted as a failed source and logged (with the
    path redacted) instead of silently looking like an empty source. The 0.5.1 note promised this;
    it now holds. The scan budget now bounds visited entries (200,000) and matches (20,000)
    separately, so non-matching session files cannot starve the real sidecars.
  • readCapped re-stats the handle after reading, so truncated means "bytes exist past the
    buffer": a session file that shrank while it was read no longer reports a stale truncation and
    loses its last complete line. A file that exists but cannot be opened (EACCES/EPERM/EISDIR)
    degrades to missing instead of failing the tool.
  • mode: "full" counts indices the manifest never offered (or that hold no text) as skipped, so a
    selection that resolves to nothing reports truncated instead of a clean empty result.
  • withoutPaths covers C:/, UNC roots, file:/// and paths containing spaces
    ("C:\Program Files...", "/home/alice/My Docs/..."). The tool details no longer carry the full
    database path, and log lines that quote raw errors/stacks pass through it as well, so the 0.5.1
    "logs have no absolute paths" note now holds for them too.
  • collectMsgIds skips a first field whose entries are all unusable instead of letting it shadow a
    valid second field.
  • A session_shutdown that is still finishing (grace period or final scan) when a new
    session_start lands no longer closes and latches the new session's store; it detects that it was
    superseded and leaves the store open.
  • configureForTests resets the cached connection, session generation, background scan and latch,
    so tests no longer depend on every test file owning its own process.
  • package.json drops the leftover publishConfig and prepublishOnly (the package is private
    and distributed through git tags), and the README update examples use the same .git URL as the
    install section.

v0.5.1

Choose a tag to compare

@github-actions github-actions released this 10 Sep 11:16

Changed

  • Distribution is git-only. The release workflow no longer publishes to npm: it checks that the
    tag matches package.json, re-checks the committed dist/, and creates the GitHub release from
    the matching CHANGELOG.md section. pi install git:... is the only documented install path,
    and the npm badge and npm install/uninstall sections were removed from both READMEs.
  • The store file is created 0600 where the OS supports it (ignored on Windows).
  • Errors returned to the model by memory_search/memory_expand no longer contain absolute local
    paths; the full trace stays in the extension log.
  • CI limits the workflow token to contents: read, the release workflow extracts release notes
    with a literal version match instead of a regular expression, and the declared
    billion-context-pi peer range is >=0.1.65 to match the documented tested version.
  • The install docs are written for the public repository: HTTPS first (no credentials required),
    SSH as the alternative, plus an "Update" section and a release badge.
  • Both workflows pin actions/checkout and actions/setup-node to commit SHAs. The repository
    also gained issue forms (bug report with version/Node/pi fields), a pull-request template,
    Dependabot configuration, and a Code of Conduct.

Added

  • README "Scope: single user, single machine" section: no accounts, server, sync, or per-user
    permissions; one store per OS user; SQLite WAL locking is single-host only, so the database file
    must never be shared; ephemeral homes rebuild the index; redaction is best-effort over a
    plaintext, permission-unhardened store; sync sidecars, not the database, across machines.

Fixed

  • memory_expand hardening after an external review of 0.5.0:
    • mode: "full" now requires a non-empty select instead of rendering the whole block,
      which contradicted the documented two-step contract.
    • expandMaxReadBytes is a real read cap again: the session file is opened and read in bounded
      chunks instead of being slurped whole and sliced afterwards.
    • A per-entry trim is reported as truncated again (a later assignment always cleared the
      flag), and the rendered text is guaranteed to stay within maxChars.
    • A missing/renamed session file degrades to missing references instead of throwing, matching
      the documented behavior.
    • custom_message entries (extension-injected context such as ACP compaction notices) are
      expandable again; unknown content-item types render an explicit [unsupported ...]
      placeholder instead of being dropped from the entry.
    • Re-ingesting a sidecar clears stored pointers the sidecar no longer lists, so expansion can no
      longer return messages that are no longer part of the block.
    • The 0.5.0 pointer migration now clears a freshly seeded watermark row, so an orphaned source
      row still gets its pointers backfilled.
    • ~ in dbPath/sourcesPath/logPath expands to the home directory instead of creating a
      literal ~ directory.
    • The ambiguous-block message reports the real number of matches instead of the page size.
    • Manifests no longer print an absolute session path, and generated acp_summary_* references
      are labelled synthetic rather than counted as recoverable text.
  • Scan and store lifecycle hardening after a follow-up review:
    • A background scan can no longer reopen the closed store after session_shutdown: the scan
      aborts per file when the session generation moved on, and getDb() refuses to open a latched
      store instead of recreating it.
    • One failing allow-list source (unreadable root, corrupt opencode DB, adapter bug) no longer
      aborts every remaining source; failures are counted and logged per source.
    • An allow-list entry with an unknown adapter is rejected and logged when the list is read,
      instead of falling through to the permissive scan-time path.
    • /memory prune writes its tombstones and deletes in one transaction, so a crash cannot leave
      the two halves inconsistent, and a failing VACUUM is logged instead of failing the prune.
    • An ingest that finishes reading a file after session_shutdown returns store closed instead
      of touching a null handle and then failing again in ROLLBACK.
    • A scan whose allow-list root holds more than 20,000 entries stops at the cap and logs it,
      instead of walking an accidentally over-broad tree to the end.
    • Log lines no longer contain absolute session or working-directory paths unless debug is on
      (the store path is logged as a file name) — logs are what users paste into public issues.
    • SECURITY.md no longer points "above" at a section that sits below it.