mirror of
https://github.com/ruvnet/ruflo.git
synced 2026-09-14 14:01:28 +08:00
docs/chatgpt-plugin-reference
1903 Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
 rUv
|
df87b0db33 |
chore(release): 3.41.2 — memory index protection, daemon consent, sibling disclosure (#3283)
#3224 — a curate may now grow or reorder MEMORY.md and never shrink it. The bridge rebuilt the index from its own topic files and overwrote whatever was there; the #1556 guard only skipped the write when nothing matched, but the bridge writes the first topic file itself, so after the first insight a hand-maintained index was replaced with a stub. A smaller generated view now goes beside the file instead, and the bridge says it declined. #3278 — autoStart:false is honoured. The disable check read claude-flow.config.json daemon.autostart while init generates .claude/settings.json claudeFlow.daemon.autoStart: a different file and a different capital S, so the setting a fresh project ships was never consulted, under a comment saying it exists to prevent unintended token consumption. #3228 (half) — retrieve, search and an empty list now disclose the sibling AgentDB store. 3.41.1 shipped that disclosure in `memory list` alone, so the other three kept returning a confident negative while the rows sat in the file they do not read. Claude-Session: https://claude.ai/code/session_013u4pmL9ZUAXb6usVQgNo67 |
||
|
rUv
|
13880e3aa4 |
fix(memory,daemon): stop destroying a user index, honour autoStart:false, and finish the sibling disclosure (#3282)
Three from the current issue backlog, all confirmed in source before touching anything, and all with a test that fails without the fix. #3224 — curateIndex() rebuilt MEMORY.md from the bridge's own topic files and overwrote whatever was there. The #1556 guard only skips the write when NOTHING matched, but the bridge writes the first topic file itself, so after the first insight the guard stops applying and the rebuild replaces a hand-maintained index with a stub: 75 curated lines and 49 links down to 6 in the reported case. The bridge only knows its own topic files, so everything else looks like nothing to it. A curate may now grow or reorder the index and never shrink it; when the generated view is smaller it is written beside the file as MEMORY.generated.md and the bridge emits index:preserved rather than going quiet. #3278 — the autostart disable check read claude-flow.config.json with the key daemon.autostart, while init generates .claude/settings.json with claudeFlow.daemon.autoStart. A different file AND a different capital S, so the one setting a fresh project actually ships was never consulted. It is written false under the comment "Opt-in only — prevents unintended token consumption (#1427, #1330)", which the runtime then ignored. Both files and both spellings are now read: someone who wrote autostart false anywhere sensible meant it. #3228 (half) — 3.41.1 disclosed the sibling AgentDB store from `memory list` only. retrieve, search and an empty list kept returning a clean negative while the rows sat in the file this interface does not read; the reported Windows install has 31,673 rows in that other store answering found:false. A confident negative is worse than an error, because nothing prompts anyone to look further. That was a gap in my own fix, not a new defect. Verified the tests are worth having by reverting each fix and re-running: #3224 fails 1/2, #3278 fails 2/4, and all six pass again restored. Claude-Session: https://claude.ai/code/session_013u4pmL9ZUAXb6usVQgNo67 |
||
|
rUv
|
cda8b04f6a |
chore(release): 3.41.1 — memory path identity and Seraphina truncation (#3273)
Two fixes, both found by someone hitting them rather than by us. #3196: an explicit database path now selects that database. The bridge cached one registry globally, so the first caller decided the file for the whole process and every later caller's --path or CLAUDE_FLOW_DB_PATH was accepted and then silently ignored. A CLI write and an MCP write landed in two files, both reporting success, and neither interface could see the other's rows. The 'memory list' command also stopped printing a confident total for one store while rows sat unread in its sibling — it now names the store it did not read. Reported downstream by pacphi/agentic-kit#213. #3267: Seraphina no longer passes a truncated reasoning dump off as coordination guidance. The routed model spent its whole token budget reasoning and never emitted its JSON, and the empty result read as "the swarm needs nothing" — the one conclusion a coordinator must never reach by accident. Claude-Session: https://claude.ai/code/session_013u4pmL9ZUAXb6usVQgNo67 |
||
|
rUv
|
e6f116b5c0 |
fix(memory): honour an explicit database path, and stop counting one store as if it were all of them (#3196) (#3271)
* fix(memory): honour an explicit database path, and stop counting one store as if it were all of them (#3196) Two writes, both reporting success, landing in two files neither interface could see whole. Reproduced on 3.41.0 in a clean directory: .swarm/memory.db CLI-written key 1, MCP-written key 0 .swarm/agentdb-memory.db CLI-written key 0, MCP-written key 1 The cause is a cache, not a path calculation. getRegistry() kept one global instance, so the first caller to touch the bridge decided the file for the whole process; every later caller's dbPath was accepted and then silently ignored, because the function returned the existing instance without ever comparing paths. `--path` and CLAUDE_FLOW_DB_PATH have always advertised that they select a database. Now they do: the cache is keyed by resolved path, so two paths in one process are two registries, and a relative path and its absolute form are one. The second half is worse than the split and is fixed separately. `memory list` printed "Showing 1 of 1 entries" — not a warning, not a partial result, a confident total describing one file as though it described the memory. A downstream integrator measured 51 rows in one store and 16,671 in its sibling, with 50 keys existing only in the smaller one. Silence would have been recoverable; a wrong total is not, because nothing prompts anyone to look further. The CLI now names the store it did not read, and how to read it. Deliberately NOT done: merging the two files. #2786 gave AgentDB its own plaintext file so native better-sqlite3 does not open an encrypted memory.db, where it fails with "file is not a database" and silently disables the learning system. Unifying the path would re-break encryption at rest. Both corpora stay, and both stay readable. The sibling count is read-only, opens nothing it cannot open read-only, and returns null when the file is missing or unreadable — an unreadable store is not evidence of an empty one, so the CLI stays silent rather than claim zero. Verified against the original reproduction: the MCP-written rows are now disclosed by name and count after the total. Three regression tests cover sibling identity, path resolution, and silence on an unreadable store. Reported-by: pacphi (pacphi/agentic-kit#213) Co-Authored-By: RuFlo <ruv@ruv.net> Claude-Session: https://claude.ai/code/session_013u4pmL9ZUAXb6usVQgNo67 * fix(memory): make the bridge test seam override every path, not two guessed keys Keying the registry cache by path broke #2968's fixture. That test installs a fake registry through __setMemoryBridgeRegistryForTests and then calls bridgeStoreEntry with its own temp dbPath, so an override installed under the two keys the seam could guess was never consulted: the call resolved a third path, missed the cache, and tried a real init instead of the fake. A test seam means 'use this registry', not 'use it if the caller passes the path I assumed'. It is now a separate override that getRegistry() returns first, kept deliberately outside the path cache so the two concerns cannot collide again. Verified: #2968 and the #3196 identity tests pass together, and the bridge dbPath, durability, purge and row-visibility suites are unchanged. Co-Authored-By: RuFlo <ruv@ruv.net> Claude-Session: https://claude.ai/code/session_013u4pmL9ZUAXb6usVQgNo67 |
||
|
ruv
|
cb10bf617e |
chore(release): 3.41.0 — open swarm federation CLI/MCP, channels, Seraphina
@claude-flow/cli, claude-flow, ruflo -> 3.41.0 (lockstep). Co-Authored-By: RuFlo <ruv@ruv.net> Claude-Session: https://claude.ai/code/session_015y2UjPSd3VX1CYnPstZJ5D |
||
|
rUv
|
90bff10f15 |
feat(x-gateway): public and private swarm channels (ADR-386) (#3261)
Coordination on x.ruv.io was one flat topic: every member read every message,
and there was no way to keep a stream separate — or confidential — without
standing up another relay tenant.
Channels are one tag on the existing events, in two visibilities:
public c=pub:<name> plaintext, any relay member reads it
private c=prv:<16 hex> NIP-44 v2 ciphertext, k=enc hides the message type
The private channel key is generated client-side and cached at
~/.ruflo/channels.json (0600). Access is granted by sealing that key to a
member's pubkey over ECDH and publishing it as a ChannelGrant only they can
open. The gateway holds no channel keys and cannot decrypt — a hosted gateway
that encrypted on your behalf would be a custodian of every private channel on
the service, which is the thing worth not building.
Not hidden, and said so in the ADR: the relay sees that a channel exists, its
opaque id, who published and when. There is no revocation either; removing
someone means rotating the channel.
The channel tag is 'c', not the obvious 'h'. buzz-relay already implements
NIP-29-style server-side group membership on : an h-tagged event publishes
fine and the matching REQ comes back CLOSED "restricted: not a channel member",
so the author cannot read back their own message. Measured every other
single-letter tag against the live relay — c/d/g/l/m/r/x/y/z are all indexed and
unrestricted.
Surfaces: gateway channel_list / channel_sync / channel_publish (public only,
admin-gated) + ruv://swarm/channels; client x_federation_channel_{create,grant,
accept,publish,read,list} and 'ruflo federation channel --action ...'.
Verified: gateway 12/12, CLI 11/11, and a live two-key E2E on wss://relay.ruv.io
— grantee reads the plaintext, an outsider key cannot decrypt, the ciphertext
does not contain the secret, and the gateway's own channel_sync returns
encrypted:true. Deployed as ruflo-x-gateway-00008-s6n.
Claude-Session: https://claude.ai/code/session_013u4pmL9ZUAXb6usVQgNo67
|
||
|
rUv
|
6f29819fe6 |
fix(metaharness): bump the darwin pin to 0.10.2 so pin-drift and clean-install go green on main (#3262)
main is red on two required gates: @metaharness/darwin is declared ~0.9.0 while npm latest is 0.10.2, so check-metaharness-pins reports STALE. The drift predates #3256 — it merged with these gates already failing — but it is on main now. Checked the bump rather than assuming it: unpacked both tarballs and compared. 0.9.0 and 0.10.2 ship the same bin (metaharness-darwin), the same exports (. and ./gepa), and the same gepa symbols this repo imports (loadCand6Genome, CAND6_GENOME_PATH). Nothing we call was removed. MH_DARWIN_PIN moves with it, since the mechanical oracle shells out to npx @metaharness/darwin@<pin> and an out-of-range pin is its own drift failure. pnpm lockfile regenerated in the same commit — v3 is a separate pnpm workspace and a stale lockfile takes the frozen-lockfile jobs down. Verified: check-metaharness-pins exits 0 with every pin reported current. Claude-Session: https://claude.ai/code/session_013u4pmL9ZUAXb6usVQgNo67 |
||
|
rUv
|
dbf450a927 |
feat(federation): ruflo CLI + MCP integration for x.ruv.io, and Seraphina (swarm queen) (#3256)
* feat(federation): ruflo CLI + MCP integration for x.ruv.io, and Seraphina (swarm queen)
Integrate the open swarm federation into ruflo itself, and add Seraphina —
a primary-coordinator / swarm-queen guidance tool in the style of the ruOS
assistant terminal (invoked from a terminal or any MCP client).
MCP tools (src/mcp-tools/x-federation-tools.ts), registered in the barrel and
the in-process registry so `callMCPTool` and the MCP server both see them:
x_federation_sync / roster / claims / registry (open reads)
x_federation_publish / invite_mint / admit (gateway-identity writes,
require RUFLO_X_ADMIN_TOKEN; fail closed with no network call)
Each description follows ADR-112 ("Use when … wrong because …").
Seraphina (src/mcp-tools/seraphina-tools.ts): `seraphina_guidance { goal }`
gathers the live roster, claims board and recent messages from the x.ruv.io
gateway, compacts them (dedupe by from|type, cap 15 — a cheap tier drowns in
repeated PeerHellos), and asks the cognitum meta-llm gateway
(https://api.cognitum.one/v1/messages, model cognitum-auto by default with a
tier override) using a queen system prompt that treats message content as
data, respects one-owner-per-resource claims, and returns
{ guidance, proposals[], risks[] }. Proposals are advisory. JSON is extracted
by slicing the outermost object so fenced/prose-wrapped answers still parse.
Key from SERAPHINA_METALLM_KEY (GCP secret seraphina-metallm-api-key).
CLI: `ruflo federation sync|roster|claims|registry|invite|admit|publish`
(src/commands/federation.ts), thin over the same tools.
Skill: .agents/skills/open-federation/SKILL.md documents CLI, tools,
Seraphina, the claims rules, the canonical NIP-42 relay-tag gotcha, and
onboarding.
Tests: 11 (x-federation: RPC shaping, SSE parse, resource mapping, fail-closed
admin gating, isError surfacing; seraphina: registration, fail-closed key,
context gathering + queen prompt + x-api-key, tier override, fenced-JSON
extraction). Project typechecks at 0 errors.
Verified live: Seraphina against the real 3-node federation returned 3
structured proposals + 4 risks via cognitum-auto (routed to a cheap tier).
Co-Authored-By: RuFlo <ruv@ruv.net>
Claude-Session: https://claude.ai/code/session_013u4pmL9ZUAXb6usVQgNo67
* fix(federation): ADR-125 env-var precedence — URL config takes a flag/arg, credentials registered env-only
The env-var-precedence audit correctly flagged the new process.env reads.
Config URLs now have a precedence path: `ruflo federation --gateway` feeds a
`gatewayUrl` tool arg (and Seraphina takes `metaLlmUrl`/`gatewayUrl`) that
wins over RUFLO_X_GATEWAY_URL / SERAPHINA_METALLM_URL, documented per ADR-125.
Credentials (RUFLO_X_ADMIN_TOKEN, SERAPHINA_METALLM_KEY) are registered as
env-only escape hatches with rationale: a secret must never be a CLI flag.
+1 test (arg precedence, arg not forwarded upstream). Audit passes locally.
Co-Authored-By: RuFlo <ruv@ruv.net>
Claude-Session: https://claude.ai/code/session_013u4pmL9ZUAXb6usVQgNo67
* feat(federation): `ruflo federation join --code` — self-service invite access with your own key
The user-facing path into the open swarm. Decentralized by design: the user
generates/holds THEIR OWN Nostr key (~/.ruflo/nostr.key, 0600), redeems an
invite code with a NIP-98-signed claim directly against the relay (no admin in
the loop), proves membership via NIP-42, and can then publish as themselves.
The gateway never signs for a user.
- src/mcp-tools/x-federation-join.ts: x_federation_join { code, relayHttp?,
relayWs?, keyFile? } — validates the code shape before any network call,
loads/creates the key, NIP-98 claim, NIP-42 verify, returns pubkey + role.
ADR-125: args take precedence over RUFLO_X_RELAY_HTTP / RUFLO_X_RELAY_WS /
RUFLO_NOSTR_KEY_FILE.
- CLI: `ruflo federation join --code v2.…`
- nostr-tools added as an optionalDependency (secp256k1/Schnorr is not in
node:crypto); the tool degrades with an install hint when absent. pnpm
lockfile regenerated in this PR (frozen-lockfile CI).
- 4 tests: 0600 key create/reuse, valid NIP-98 header (kind 27235, verifies,
bound to url+method+payload hash), malformed code rejected pre-network,
ADR-112 description. Suite: 16/16; tsc 0; env-var audit passes.
Co-Authored-By: RuFlo <ruv@ruv.net>
Claude-Session: https://claude.ai/code/session_013u4pmL9ZUAXb6usVQgNo67
* fix(federation): default relay -> wss://relay.ruv.io; join test tolerates missing nostr-tools
- x-federation-join defaults now point at the canonical relay.ruv.io host (Cloud Run
domain mapping for buzz-relay); the raw run.app host stays routable for old clients.
- Validate the invite-code shape before the optional-dependency check so a bad code
fails fast whether or not nostr-tools is installed.
- The root Test Suite runs the CLI tests via root npm ci, which never installs the
CLI's optionalDependencies: the crypto cases are it.skipIf(!nostr-tools) so the
file no longer trips the CI test ratchet.
Co-Authored-By: RuFlo <ruv@ruv.net>
Claude-Session: https://claude.ai/code/session_013u4pmL9ZUAXb6usVQgNo67
|
||
|
ruv
|
2629de628b |
chore(release): 3.39.3 -> 3.40.0
Ship cross-host federation (agentbbs) + claims. New MCP tools:
federation_bbs_{identity,peer_add,peers,serve,register,publish,sync,watch}.
Signed Ed25519 envelopes, registry-anchored pinning, HTTP pull transport.
Includes the memory-search recall fix (#3252).
Co-Authored-By: RuFlo <ruv@ruv.net>
Claude-Session: https://claude.ai/code/session_013u4pmL9ZUAXb6usVQgNo67
|
||
|
rUv
|
e111bfbd7f |
fix(memory,tests): restore memory search recall and green the CLI suite (69 → 0) (#3252)
* fix(memory,tests): restore `memory search` recall and green the CLI suite (69 → 0) The CLI suite had 69 failing tests across 50 files. Root-causing each rather than adjusting expectations turned up one real product regression, two real bugs, and a set of tests asserting things about the machine instead of the code. PRODUCT REGRESSION — `memory search` recalled nothing (re-break of #2558) `memory search` returned zero results for content that matched word for word, while store/list/retrieve worked. Bisected to the threshold default: * 2026-07-04 #2558 restored recall, designed around a 0.3 threshold. Its fusion scores a full-coverage keyword hit as 0.6*max(0,semantic) + 0.4*lexical, so with a non-positive cosine — routine for a one-word query — a PERFECT keyword match tops out at exactly 0.40. * 2026-07-26 #2790 set the CLI default to 0.7. Above 0.4, keyword recall is mathematically unreachable, so #2558 silently regressed. Instrumented the live path to confirm before changing anything: key=note/alpha terms=["connectivity"] cov=1 bm25=0.029 sem=-0.883 lex=1.000 score=0.400 thr=0.7 → dropped Default restored to 0.3 at both sites, with the ceiling documented so it is not raised past 0.4 again. Verified end to end: a shared keyword recalls all three entries, a unique keyword recalls only its own. This was broken on main too. REAL BUG — anchor containment rejected any project under a symlink containedPath() compared a realpath'd root against a NON-realpath'd candidate, so on macOS (/tmp → /private/tmp) every project looked like an escape. Now compares like with like. Both guarantees re-verified by probe: ../ traversal, deep traversal, absolute-outside and symlink-escape all still rejected; symlinked-root spelling now accepted. REAL BUG — a test that could not fail funnel.test.ts asserted `elapsed < 100ms` as a proxy for "no network call" — its own comment conceded a future fetch() would still pass. It also failed intermittently (118ms) under full-suite CPU contention. Replaced with an actual assertion on fetch/http.request/https.request, patched via the CJS copies (ESM namespaces are non-writable). Negative-controlled: the technique observes a real call and restores cleanly. TESTS THAT MEASURED THE MACHINE, NOT THE CODE * 51 wasm tests asserted `available === true` for optional packages that were simply not installed, gated only on `process.env.CI`. Now also gated on real availability. The 7 tests in those files that need no WASM were lifted into their own blocks so they keep running rather than being skipped along with the rest — no coverage traded away for a green tick. * 4 agenticow tests lacked the `skipIf(!havePkg)` guard their 4 siblings had; in degraded mode the verb is inert and returns before validation. * policy-runtime hashed the raw tmpdir path while the source hashes the CANONICAL one, so it looked for a trust anchor that is never written there — and its cleanup was deleting a nonexistent path, leaking real anchors into ~/.config on every run. * sona's provider allowlist predated the ruvector / wasm-embedder / @claude-flow/embeddings backends. * ruvector/index asserted vi.mock satisfies a *dynamic* import() of a bare specifier. It does not under vitest 4 — the specifier is rewritten to '/@id/@ruvector/core' and fails with ERR_MODULE_NOT_FOUND. Now pins the contract the try/catch actually provides: unloadable → false, never throws. TIMEOUTS — the "flakiness" was a fixed 5s budget Six memory/intelligence files failed with "Test timed out in 5000ms", never an assertion. They initialise a real ONNX embedder and SQLite; alone they are fast, but the full suite saturates every core (~440% CPU). Because it depended on scheduling, a different file failed each run. Raised testTimeout/hookTimeout to 30s. (Process-level isolation was tried first and ruled out — forks changed nothing, which is what identified the cause as time, not shared state.) Verified: 3 consecutive full runs at 0 failed / 3572 passed on the feature base, and 0 failed / 3537 passed on this branch's base. Co-Authored-By: RuFlo <ruv@ruv.net> Claude-Session: https://claude.ai/code/session_013u4pmL9ZUAXb6usVQgNo67 * fix(tests): make ruvector availability assertion environment-independent CI caught what my first pass missed. The rewritten test asserted `isRuvectorAvailable()` resolves to `false`, but whether the dynamic `import('@ruvector/core')` resolves is environment-dependent: - under vitest's module runner a bare-specifier dynamic import is rewritten to '/@id/@ruvector/core' and fails (ERR_MODULE_NOT_FOUND) → false → my `toBe(false)` passed locally - a normal install where the package resolves → true → `toBe(false)` fails, which is exactly what the CI test-ratchet flagged The prior `toBe(true)` had the same defect in the other direction. Pin the invariant that holds in BOTH environments instead — the only thing the try/catch actually promises: it resolves to a boolean and never throws. No specific resolution outcome is asserted. Verified 33/33 locally; the sibling "should return boolean" test already covers the type, so this now documents the no-throw contract distinctly. Co-Authored-By: RuFlo <ruv@ruv.net> Claude-Session: https://claude.ai/code/session_013u4pmL9ZUAXb6usVQgNo67 |
||
|
rUv
|
72df96c297 |
feat(agentbbs): cross-host federation — signed envelopes, pinned peers, union merge (#3251)
* feat(agentbbs): cross-host federation — signed envelopes, pinned peers, union merge
Phase 1 gave every host a local append-only room log and derived roomId
deterministically from the room label, so two hosts that register #sales
already compute the same roomId without talking. This adds the layer that
actually moves envelopes between them.
Design. A room log is an append-only set of immutable envelopes, so reconciling
two hosts is a set union — there is no conflicting write to arbitrate and
nothing for a consensus round to decide. Union is commutative, associative and
idempotent, so sync is order independent and safe to retry. That is a
grow-only set keyed on envelopeId, and it is cheaper and less failure-prone
than the Byzantine agreement the plugin README gestures at.
What union does not give you is authenticity: if any peer can inject, the merge
faithfully replicates forgeries. So each host now holds a persistent Ed25519
identity (Phase 1's key was ephemeral per process, which no peer can pin),
every published envelope is signed, and a receiver verifies against the key it
pinned at peer-add time rather than one carried in the envelope — otherwise an
attacker signs with their own key and claims any origin.
Transport is pull-based HTTP: peers poll GET /agentbbs/v1/rooms/:roomId/
envelopes?since=N. Pull needs no inbound connectivity, tolerates a node being
offline, and leaves ingest volume under the receiver's control; push would make
every node an unauthenticated write target. The server binds 127.0.0.1 unless
a bindHost is passed explicitly, and exposes no route that mutates state.
Bounds are enforced receive-side where a sender cannot negotiate them away:
envelopes per sync, bytes per envelope, bytes per response, peer count, and a
hop limit that is incremented on merge so a cycle terminates. hops is excluded
from the signed material, since it is mutated in transit by design.
Five tools: identity, peer_add, peers, serve, sync.
Tested: 35 new tests, most of them the trust boundary — forged payload,
forged origin, key substitution, unpinned signer, cross-room injection, replay,
oversize, hop exhaustion, traversal at the HTTP boundary, and private key never
served. Convergence and idempotence are asserted against two real nodes over a
real socket, not mocks. Verified end to end with three independent nodes:
all three converge on the same envelope set, a second round merges nothing,
and an unpinned node's injection is rejected.
The Phase 1 structural test pinned the surface at exactly 4 tools; it now pins
9 and names both groups, so adding a tool stays a deliberate contract change.
Full package suite: 68 failures against 70 on main, +35 passing — no
regressions.
Co-Authored-By: RuFlo <ruv@ruv.net>
Claude-Session: https://claude.ai/code/session_013u4pmL9ZUAXb6usVQgNo67
* ci(agentbbs): stop rule 2 firing on the federation wire namespace
ADR-164 rule 2 exists to stop `agentbbs` becoming a mandatory dependency: a
static `import ... from 'agentbbs'` is forbidden, and any other mention has to
sit behind loadAgentbbs / a dynamic import. Its catch-all is a substring test
for "agentbbs" anywhere in the file.
agentbbs-federation.ts trips that catch-all without importing anything. Its
only remaining mentions after comment-stripping are string literals that never
reach a module resolver: the '/agentbbs/v1/...' HTTP route prefix and the
'agentbbs:<kind>:' hash domain-separators.
Adds those two shapes to the existing benign-pattern strip list. The
static-import check is untouched and remains the real gate.
Verified this narrows the check rather than loosening it, by running the
workflow's own rule-2 body verbatim against fixtures:
- the real federation module -> passes (false positive gone)
- `import { x } from 'agentbbs'` -> still caught
- unguarded runtime reference -> still caught
- guarded `await import('agentbbs')` -> passes
Co-Authored-By: RuFlo <ruv@ruv.net>
Claude-Session: https://claude.ai/code/session_013u4pmL9ZUAXb6usVQgNo67
* ci(agentbbs): feed rule 2 via a quoted heredoc so bash stops mangling it
The rule 2 fix in
|
||
|
ruvnet
|
e855ae2e4e |
chore(release): 3.39.2 -> 3.39.3
Ships #3250: the agentbbs MCP tools gated on `await import('agentbbs')`, but the published package is a CLI-only launcher with no importable entry point, so all four federation_bbs_* tools reported degraded regardless of install state. Replaced with a subprocess probe. Includes the follow-up hardening: the probe's shell is now gated to win32 instead of always on. The probed binary comes from AGENTBBS_BIN, and under a shell that argument is command-interpreted, so a crafted value executed a trailing command (verified locally). POSIX now spawns without a shell. Co-Authored-By: RuFlo <ruv@ruv.net> Claude-Session: https://claude.ai/code/session_013u4pmL9ZUAXb6usVQgNo67 |
||
|
rUv
|
07f1cf3a3f |
fix(agentbbs-tools): detect the real CLI-only agentbbs package instead of a nonexistent JS import (#3250)
* fix(agentbbs-tools): detect the real CLI-only agentbbs package instead of a nonexistent JS import
The published `agentbbs` package (0.2.1) is a CLI-only launcher — a `bin`
script with no importable JS entry point — that downloads/builds a Rust
binary and shells out to it. loadAgentbbs()'s `await import('agentbbs')`
could therefore never resolve, even with the CLI correctly installed and
on PATH, so all 4 federation_bbs_* tools reported degraded:true
unconditionally regardless of install state.
Replace the dynamic import with a subprocess presence probe
(agentbbsCliAvailable(), `agentbbs --version`), matching how the rest of
the codebase treats optional CLI tools. shell:true is required for the
probe to find npm's Windows .cmd shim (child_process does not consult
PATHEXT otherwise).
Making the probe actually succeed surfaced two further pre-existing bugs
in code that was previously always skipped:
- vitest.config.ts externalized @noble/ed25519 alongside genuinely-optional
deps (agentic-flow, agentdb, ...), but it is an always-installed hard
dependency; externalizing it broke dynamic-import resolution under Vite's
SSR transform.
- federation_bbs_human_join dynamically imported @noble/ed25519 a second
time (getSigningKey() already does), which is now deduplicated behind one
memoized loadEd25519() loader.
All 12 tests in agentbbs-tools.test.ts now pass, including the 4
happy-path tests that were previously always skipped.
* fix(agentbbs-tools): gate the probe shell to win32 instead of always on
`shell: true` was unconditional, and the probed binary comes from
`process.env.AGENTBBS_BIN`. Under a shell that argument is command-interpreted,
so `AGENTBBS_BIN='agentbbs --version; touch /tmp/PWNED'` executes the trailing
command. Verified locally: the file is created with `shell: true` and the spawn
fails with ENOENT once the shell is gated off.
This is the case browser-tools.ts already warns about in the comment beside the
same pattern -- "if user-controlled args are ever added, escape them before
spawn" -- and `shell: process.platform === 'win32'` is the convention already
used at browser-tools.ts:49, commands/init.ts:856 and :1194, and
init/helpers-generator.ts:1459. Unconditional `shell: true` appeared nowhere
else in the package.
The Windows behaviour the original was reaching for is unchanged: cmd.exe still
resolves the `agentbbs.cmd` shim via PATHEXT. POSIX simply stops interpreting
the override. Added windowsHide to match the sibling call sites.
Verified the fix still does what it set out to do: with the CLI resolvable the
suite is 12/12 and the four previously-always-skipped happy-path tests execute,
with the shell disabled on POSIX.
Co-Authored-By: RuFlo <ruv@ruv.net>
Claude-Session: https://claude.ai/code/session_013u4pmL9ZUAXb6usVQgNo67
|
||
|
rUv
|
498a238799 |
fix(security): make the fast-uri CVE fix actually reach shipped packages (#3222) — 3.39.2 (#3247)
* fix(security): make the fast-uri CVE fix actually reach shipped packages (#3222) 3.39.1 bumped fast-uri only in the ROOT package.json. Post-publish validation caught that this never reached users: a fresh `npm install ruflo@3.39.1` still resolved fast-uri@3.1.5, squarely inside the advisory range (GHSA-5jgf-p345-68v8, high, CVSS 7.5, >=3.1.3 <3.1.6). Root cause: v3/@claude-flow/cli/package.json carried its own hard pin "fast-uri": "3.1.5" in dependencies, and that is the package that actually ships. The root bump did not override it -- the #2112 lesson recorded in CLAUDE.md, that root overrides do NOT propagate to the published ruflo wrapper, applies to the CLI package too. - v3/@claude-flow/cli: fast-uri 3.1.5 -> ^3.1.6 (the shipping dependency) - ruflo: add "fast-uri": ">=3.1.6" override so the wrapper cannot regress independently of the CLI Verified by fresh-installing the published tarball rather than trusting the lockfile: that is what surfaced the gap in the first place. Refs #3222, #2112 Co-Authored-By: RuFlo <ruv@ruv.net> Claude-Session: https://claude.ai/code/session_01JJnDvpYYk2rTvjkudQUoCk * fix(ci): refresh v3 pnpm lockfile for the fast-uri bump The 3.39.2 fast-uri change touched v3/@claude-flow/cli/package.json but only the npm lockfile was regenerated. v3/ is a separate pnpm workspace and CI runs pnpm install --frozen-lockfile with working-directory: v3, which rejected the mismatch (ERR_PNPM_OUTDATED_LOCKFILE) and cascaded into 34 failing checks -- every one an install failure, not a real regression. Diff is 6 lines, fast-uri only: specifier 3.1.5 -> ^3.1.6, resolved 3.1.7. Verified pnpm install --frozen-lockfile now exits 0. Co-Authored-By: RuFlo <ruv@ruv.net> Claude-Session: https://claude.ai/code/session_01JJnDvpYYk2rTvjkudQUoCk |
||
|
rUv
|
d55b1bfeac |
fix(memory): stop reporting a gated native bridge as the active backend (#3228) + fast-uri CVE (#3245)
* fix(memory): stop reporting a gated native bridge as the active backend (#3228) On Windows the #3024 kill-switch makes getRegistry() return null, so every memory read/write silently falls back to the sql.js store (memory.db) instead of the AgentDB corpus (agentdb-memory.db) the bridge would have opened. The reporter measured the consequence: a canary store/retrieve round-trip SUCCEEDS because both halves use the same unintended file, while an existing live-store key returns found:false against 31,673 real rows. The status surface hid the substitution. getHNSWStatus() gated its bridge branch on whether the bridge MODULE was loaded -- which it is, even when the registry is gated off -- so describeBackend() printed "sqlite (bridge, brute-force cosine)" for a store the bridge never touched. Gate on whether the bridge is permitted to be the active path rather than on whether the module resolved. shouldDisableNativeBridge() is sync and a pure function of platform + env, so it is a faithful discriminator and needs no warm registry. describeBackend() now names the gate and its reason, so an upgrade that changes which file receives writes says so. This does not fix the opt-in allocation abort (#2948) or reroute the fallback to the AgentDB corpus; both need Windows validation. It makes the silent case loud, which is what the report asks for first. Also bumps fast-uri 3.1.5 -> ^3.1.6 (resolves 3.1.7), clearing GHSA-5jgf-p345-68v8 (high, host confusion via skipped IDN canonicalization) and the ajv high that inherited it. Same change as #3222 by @aeonframework, applied here because CI was never authorized to run on that PR. Refs #3228, #3222, #2948, #3024 Co-Authored-By: RuFlo <ruv@ruv.net> Claude-Session: https://claude.ai/code/session_01JJnDvpYYk2rTvjkudQUoCk * chore(release): 3.39.0 -> 3.39.1 Co-Authored-By: RuFlo <ruv@ruv.net> Claude-Session: https://claude.ai/code/session_01JJnDvpYYk2rTvjkudQUoCk |
||
|
ruv
|
e341ec8c4a |
chore(release): 3.38.23 -> 3.39.0
Ships the ADR-322C receipt number-domain fix (#3229 / #3235). MINOR rather than patch: receipt output changes shape (fractional candidatePolicy values become scale-12 decimal strings) and policySchemaVersion moves v1 -> v2, so pre-fix receipts refuse promotion with 'policy schema changed'. Not MAJOR because no CLI surface or MCP tool signature changed. Existing signed receipts stay valid as history and are deliberately NOT migrated — re-encoding changes the content, hence the ID, hence invalidates the signature. Re-establish a champion through the explicit reset path, which already requires confirmation and a recorded reason. Validated end to end against the built dist before publishing: a receipt carrying the exact candidatePolicy from #3229 encodes to {"alpha":"0.3","subjectWeight":1,"mmrLambda":"0.5","bodyWeight":"1.5", "typePenaltyFactor":"0.5"}, contains zero fractional JSON numbers anywhere in the payload, and verifies VALID. Co-Authored-By: RuFlo <ruv@ruv.net> Claude-Session: https://claude.ai/code/session_01J2s8oQsAZjJtJ6inoYRFPj |
||
|
rUv
|
ea2df4726c |
fix(flywheel): encode candidatePolicy fractions as decimal strings (#3229) (#3235)
`ruflo metaharness flywheel run` wrote receipts whose candidatePolicy
carried binary floats — {"alpha": 0.3, "mmrLambda": 0.5} — contradicting
the contract each receipt claims to satisfy:
witness-receipt-contract.md rule 2 "Fractional values are decimal
strings, never binary floats"
conformance-checklist A3 "Every fractional value is a
canonical decimal string"
ADR-322C L22 "Signed policy fractions use
schema-quantized decimal strings"
The spec's own example strings it: {"hnswEf": 128, "hybridWeight": "0.65"}.
Confirmed in the PUBLISHED 3.38.23 artifact, not just source:
dist/src/services/harness-flywheel.js:219 `candidatePolicy: candidate`.
Two releases shipped after the 3.38.21 report without fixing it.
Root cause was two-part, and the second half is what let it survive:
1. Producer: `candidate as unknown as Record<string, unknown>` — a DOUBLE
cast (a single one will not compile) that erased RetrievalConfig's
`number` fields and laundered floats into the receipt.
2. Verifier: `assertJsonValue` accepted any finite non--0 number. That was
its entire number rule, so ruflo verified its own non-conforming
receipts while autogenous's stricter verifier correctly rejected them
(ruvnet/autogenous#15). The JSON Schema pins `decimalString` on ten
named fields and cannot express it inside `candidatePolicy`, which is
opaque by design — so the rule lived only in prose.
Fix:
- `encodePolicyFractions` encodes non-integers to scale-12 decimal strings
and leaves integers as JSON numbers, matching the contract's example.
Recurses through nested objects and arrays.
- `policyCandidateId` encodes at the hashing boundary, so the content ID is
always over the canonical form. Encoding is idempotent, so
policyCandidateId(raw) === policyCandidateId(encoded) and `verify` can
still recompute the ID from the payload.
- `assertReceiptNumberDomain` enforces the rule at the RECEIPT boundary, on
produce (before signing) and on verify. This is the half that closes the
class: without it the next writer drifts again invisibly.
- The producer's type-erasing double cast is gone.
- policySchemaVersion v1 -> v2, because fixing the encoding changes the
candidate content ID. Pre-fix receipts now refuse promotion with an
ACCURATE 'policy schema changed' instead of a confusing 'stale baseline'
hash mismatch. Signed receipts are deliberately NOT migrated: re-encoding
changes the content, hence the ID, hence invalidates the signature — a
migration would mean minting new receipts claiming to be old ones. They
stay valid as history and unpromotable; the champion is re-established
through the existing confirm-and-reason reset path.
Scope note worth keeping: a first attempt put the rule inside
`canonicalizeJcs` and broke the proposer envelope and the promotion ledger,
which share that canonicalizer and which the contract does not govern. The
full suite caught it; a test now pins that the shared canonicalizer stays
permissive.
Mutation-verified: disabling the encoder fails 16 tests, disabling the
enforcement fails 1, and dropping the boundary encoding in
policyCandidateId fails 13.
Measured A/B on one installed+built tree — strictly better, nothing broken:
baseline 40 files / 83 tests failed, 2801 passed
with fix 39 files / 82 tests failed, 2811 passed
The remaining failures are @ruvector/*-wasm optional native deps absent on
this host, present identically on origin/main.
Claude-Session: https://claude.ai/code/session_01J2s8oQsAZjJtJ6inoYRFPj
|
||
|
ruvnet
|
384b923986 |
chore(release): 3.38.21 -> 3.38.23
Patch release covering the six commits landed since 3.38.21: - #3221 gate EnhancedModelRouter's forwarded modelId on tier match - #3169 wire embedding-cosine into SmartRetrieval's MMR step - #3184 wire real cold-start measurement, drop the fabricated V2-vs-V3 benchmark - #3177 bound two ReDoS patterns, stateless PII regexes, O(P) confidence pass - #3204 fix two stale main-tree Test Suite failures - #3203 add v3/@claude-flow/mcp to the root npm workspaces 3.38.22 is intentionally skipped: @claude-flow/cli's publish for that version stuck in npm's "staged" state (published umbrellas, cli 404) and the registry refused to republish over it even after unpublish, so all three packages moved to .23 together to restore version lockstep. Co-Authored-By: RuFlo <ruv@ruv.net> Claude-Session: https://claude.ai/code/session_01Ff2xRKvYrqXJhefvcapfE1 |
||
|
rUv
|
6acc689150 |
dream(intelligence): #3220 gate EnhancedModelRouter's forwarded modelId on tier match (evaluated, ACCEPT-scoped) (#3221)
Merged after independent re-verification: 109/109 CI green, the earlier '168 failures' finding traced to an unbuilt-sibling-package local artifact (unrelated to changed files), and a subsequent CI flake confirmed unrelated via clean re-run. Overriding the PR's self-merge gate on explicit user (ruvnet) authorization. Co-Authored-By: RuFlo <ruv@ruv.net> Claude-Session: https://claude.ai/code/session_013XaRQntabVwGT9SvnW8cKK |
||
|
rUv
|
277c7bc03a |
dream(memory): #3168 wire embedding-cosine into SmartRetrieval's MMR step (evaluated, ACCEPT) (#3169)
* dream(memory): wire embedding-cosine into SmartRetrieval's MMR step (evaluated, ACCEPT) smart-retrieval.ts's MMR diversity re-ranking used token-Jaccard text overlap as its "similarity to already-selected" term, even though memory-initializer.ts's searchEntries() already computes real ONNX-embedding cosine similarity for the primary relevance score and discards it before returning. 2025-2026 production practice (LangChain's reference maximal_marginal_relevance, Qdrant's native Mmr query shipped Sept 2025, Weaviate's MMR reranker shipped in 1.37) universally uses embedding-cosine for this term; token overlap misses low-token-overlap paraphrases embeddings correctly flag as near-duplicates. Threads the embedding through: searchEntries()'s two live compute sites (RaBitQ rerank, brute-force SQL) now return it instead of discarding it; the two CLI/MCP bridge mappers (memory-tools.ts, commands/memory.ts) pass it through; SearchCandidate gains an optional `embedding` field; mmrRerank uses cosine when both candidates have one, falling back to token-Jaccard otherwise (existing callers/tests without embeddings are unaffected). controller-registry.ts's separate toCands()/applyMMR path had the same gap for a different reason (embedding available on r.entry.embedding but never copied) — fixed the same way. Evaluated via deterministic Vitest, zero LLM calls, $0 cost. New discriminating test: a low-token-overlap paraphrase (near-1.0 cosine to the seed) vs. a genuinely different topic (near-0 cosine). Baseline (git stash-isolated): picks the paraphrase second (Jaccard can't tell "low overlap because paraphrase" from "low overlap because unrelated"). Candidate: correctly picks the different topic second. Full @claude-flow/memory suite: 460/461 passing identically with/without the candidate (1 pre-existing chmod-based environmental failure, same class documented since 2026-08-15). tsc --noEmit: zero new errors (verified @claude-flow/cli's 455 pre-existing errors, from unbuilt workspace packages, are identical with and without this diff). Darwin: skipped, scope mismatch (correctness/similarity-metric fix with a discriminating unit test, not a continuous parameter with a gold-labeled corpus). Co-Authored-By: RuFlo <ruv@ruv.net> Claude-Session: https://claude.ai/code/session_01Qho5K4p5ACBt4yzCWjkBGe * dream(memory): address review remediation — split CI, real benchmark, hardening Response to ruvnet's REJECT review on #3169: 1. Split — dream-cycle-backlog-guard.yml moved out to #3205; this branch now carries only the MMR fix, rebased onto main (which already includes the merged #3203 CI fix). 2. Real benchmark (new src/mmr-benchmark.test.ts): 18-doc/6-topic labeled corpus (1 canonical + 2 low-token-overlap paraphrases per topic), run through the actual smartSearch/mmrRerank code, swept across mmrLambda in {0.3, 0.5, 0.7, 0.9} rather than one favorable value. Honest finding: at diversity-heavy lambda, candidate's recall/nDCG drop sharply vs. baseline, because cosine correctly suppresses same-topic near-duplicates that Jaccard fails to detect at all (and so "accidentally" keeps full recall). At smart-retrieval.ts's own default (lambda=0.7), recall@6/nDCG@6 are identical to baseline (1.000/0.777 both) while topic-diversity improves 0.472->0.667 and ground-truth duplicate-rate drops 0.289->0.200. Reports latency (sub-ms both paths at this corpus size) and a memory-overhead estimate. 3. Malformed/dimension-mismatch: found and fixed a real bug while writing these cases. A NaN/Infinity-poisoned embedding component reached cosineSimilarity unguarded; the poisoned NaN score meant `mmr > bestMmr` was never true (NaN comparisons are always false in JS), which could silently break MMR's selection loop and under-fill results below the requested limit. Added isWellFormedEmbedding() so malformed/empty embeddings fall back to token-Jaccard, same as absent/mismatched ones. 3 new cases (NaN/Infinity, empty array, explicit dimension-mismatch) plus a test mirroring the two real CLI/MCP output whitelists (memory-tools.ts:559-569, commands/memory.ts:633-639) to assert `embedding` cannot appear in either shape. 4. Determinism: new test runs the same input through smartSearch 5 times concurrently, asserts byte-identical result ordering every time, including a genuine same-embedding tie. Full @claude-flow/memory suite: 465/466 (1 pre-existing, unrelated environmental failure). tsc --noEmit: zero errors. Co-Authored-By: RuFlo <ruv@ruv.net> Claude-Session: https://claude.ai/code/session_01Qho5K4p5ACBt4yzCWjkBGe --------- Co-authored-by: Claude <noreply@anthropic.com> |
||
|
rUv
|
66e4a12366 |
test/hooks: fix two stale main-tree Test Suite failures exposed by #3203 (embedding width; hook-handler exports) (#3204)
* test(context-hook): assert the hash embedding width is EMBEDDING_DIM (384), not the legacy 768 tests/context-persistence-hook.test.mjs still expected createHashEmbedding to return 768 values; the hook has defaulted to 384 (the ONNX all-MiniLM-L6-v2 width, EMBEDDING_DIM) since February so hash-fallback blobs stay comparable with ONNX vectors. The mismatch was hidden while the Test Suite job died at ETARGET (fixed in #3203) and surfaced on the first PR runs after it (#3177, #3184). Assert EMBEDDING_DIM and pin it to 384; fix the stale "768-dim" docstring on the blob store. node --test tests/context-persistence-hook.test.mjs: 66 pass, 0 fail. Co-Authored-By: RuFlo <ruv@ruv.net> Claude-Session: https://claude.ai/code/session_019xHM4rAH4aaShb4DTr1n6s * fix(hooks): restore hook-handler's require.main guard and test exports dropped by the July helper sync tests/hook-handler-runwithtimeout.test.cjs failed with "runWithTimeout is not a function" on the first Test Suite runs after #3203 (#3177, #3184). |
||
|
ruv
|
db4991967c |
chore(release): 3.38.20 -> 3.38.21
Publishes the #3155 fix (fix(memory): stop seeding the bridge's ControllerRegistry with the sql.js dbPath, PR #3156) and the CI-fixing PR #3059 (agentic-flow-agent duration-assertion flake) to npm. Co-Authored-By: RuFlo <ruv@ruv.net> Claude-Session: https://claude.ai/code/session_011N1hncQ1p4pVt15q2VqaQD |
||
|
ruv
|
d413297dde |
fix(memory): stop seeding the bridge's ControllerRegistry with the sql.js dbPath
Fixes #3155. `initializeMemoryDatabase()` passed its just-resolved sql.js-facing `memory.db` path straight through to `activateControllerRegistry()` -> `bridge.getControllerRegistry(dbPath)`. That seeded the process-wide ControllerRegistry singleton with `memory.db` as AgentDB's own native better-sqlite3 database, instead of the dedicated `agentdb-memory.db` sibling `getAgentDbPath()` exists specifically to provide (#2786). Because this activation only runs on a database's first-ever init (an already-initialized `.swarm/` skips it via the #1791.6 idempotent no-op branch), a bridge started against a brand-new directory would activate against `memory.db` and work fine for that process's lifetime. A restarted bridge process, whose `.swarm/memory.db` already exists, would skip that call entirely, and its first real bridge operation would activate the registry against `agentdb-memory.db` instead (via the `dbPath || getAgentDbPath()` fallback in memory-bridge.ts). That file never received the earlier writes, so every read after a restart came back `found:false` / an empty list, silently, with no error anywhere. `activateControllerRegistry()` no longer forwards the sql.js dbPath to the bridge at all — it calls `bridge.getControllerRegistry()` with no argument, so both the init-time warm-up and every later bridge call (bridgeStoreEntry, bridgeGetEntry, ...) resolve the same file via `getAgentDbPath()`, regardless of whether this is a fresh directory or a restart against an already-initialized one. Co-Authored-By: claude-flow <ruv@ruv.net> Claude-Session: https://claude.ai/code/session_01MND33UDrtWPRH851k9w8M3 |
||
|
Dragan Spiridonov
|
36740e4304 | fix(proxy): verify effective daemon on install | ||
|
ruv
|
e21aa352fd |
chore(release): 3.38.19 -> 3.38.20
Publishes PR #3092 (fix(statusline): stop pinning intelligence to a hardcoded 0%). Co-Authored-By: RuFlo <ruv@ruv.net> Claude-Session: https://claude.ai/code/session_01BGiC4SoXiGcUHxs4TsFCeh |
||
|
rUv
|
a0b39b9dd0 |
fix(statusline): stop pinning intelligence to a hardcoded 0% (#3092)
Closes #3091. `intelligencePct` had exactly one assignment in the whole file, inside `buildLocalFallback()`. Everywhere else it was only read. So the segment showed either whatever the CLI returned or a literal 0 -- and since `applyLocalOverlays()` repairs adrs/agentdb/tests/hooks/integration/security from disk but never touched `system`, any CLI failure left the brain at 0% sitting beside segments that were still live. Nothing distinguished the two, so an absent measurement rendered exactly like a real one. That is reachable from ordinary conditions, not just exotic ones. Both candidates fail together on a normal setup: Claude Code's plugin marketplace git-clones the repo with no install step, so `bin/cli.js` exists while `@claude-flow/cli-core` does not (the case resolveCliBinCandidates() already documents), and the npx backstop dies with `npm error Invalid Version:` in any npm-workspaces root holding a member with no version field. Three changes: - `getLocalIntelligence()` joins the overlay set, deriving the percentage from `.claude-flow/neural/patterns.json` (project, then home) with the formula the sibling statusline.js already used. A CLI value greater than zero still wins. - `countPatternsCached()` keys the parse on the store's mtime+size. The store is routinely large -- 14.5MB / 1277 patterns on the machine this was diagnosed on -- and the statusline re-renders every prompt, so parsing it each time was not an option. Measured: 57.2ms cold, 0.2ms warm. - Unknown now renders as `—` rather than `0%`, and the fallback constant is `null` instead of `0`. A metric that silently pins to zero when a subprocess dies is indistinguishable from a metric that is genuinely zero, which is what made this read as a display bug rather than an outage. This is the same failure `getLocalAgentDB()` already carries a comment about -- "the statusline showed Vectors 0 despite thousands of real vectors" -- in a different segment. Applied to both committed copies. The CLI package's copy is the generator's single source of truth (#2679) and the one the drift guard byte-compares, so editing it keeps generator output and artifact in step by construction; the repo-root copy is CRLF and its endings are preserved. Verified by execution, both copies: 1277-pattern store -> 100% (previously 0%) no store anywhere -> — (previously 0%) Claude-Session: https://claude.ai/code/session_016QSCkKnxDjqU49NVVpWMK5 |
||
|
ruv
|
68a9c9f6a4 |
chore(release): 3.38.18 -> 3.38.19 (supersedes broken 3.38.17/3.38.18)
A subagent I spawned earlier (intended only to relay a status message) went out of scope and independently ran its own full fix/verify/publish cycle in parallel with mine on this same checkout, racing my edits. Between its publish and a set of stale-cached `claude-flow daemon` background processes self-healing package.json mid-write (#3005), @claude-flow/cli was published twice in a row (3.38.17, 3.38.18) with a broken `@claude-flow/mcp` dependency pin, and 3.38.18 additionally shipped entirely missing the `@claude-flow/memory` dependency. claude-flow and ruflo were published at 3.38.18 too, inheriting the same defect. 3.38.19 is the verified-correct cut: @claude-flow/mcp repinned to ^3.0.0-alpha.9 (published, has the #2990 fixes), @claude-flow/memory present at ^3.0.0-alpha.23 (has the #2977 fix). Deprecating 3.38.17/3.38.18 on npm as a follow-up; this commit does not touch the doc/metadata files that raced independently of the version bump. Co-Authored-By: RuFlo <ruv@ruv.net> Claude-Session: https://claude.ai/code/session_018ZkgSrVRWNMcXmaFiMZ7Z8 |
||
|
ruv
|
127a6546c4 |
chore(release): 3.38.17 -> 3.38.18, @claude-flow/memory 3.0.0-alpha.22 -> alpha.23
3.38.17 was published by a concurrent session mid-race with a stale
package.json (the @claude-flow/mcp pin had been reverted to the broken
3.0.0-alpha.10 by a background helper-refresh process just before publish;
see #3005). npm forbids republishing over an existing version, so cutting
3.38.18 with the verified-correct pin instead of trying to fix 3.38.17.
Also discovered @claude-flow/memory is NOT in @claude-flow/cli's
bundleDependencies (unlike @claude-flow/mcp/codex/security/plugin-agent-federation)
-- it's resolved from the registry via a normal ^3.0.0-alpha.22 semver range,
and that exact version is what's currently published, without the #2977 fix
from
|
||
|
ruv
|
c313276442 |
docs: correct MCP-server-registration claims, refresh catalog/helpers metadata (retry)
Same content as
|
||
|
ruv
|
b5b12a751a |
docs: correct MCP-server-registration claims, refresh catalog/helpers metadata
- README (root + ruflo/): the "no MCP server on plugin-only install" claim was inaccurate for ruflo-core, which ships its own .mcp.json; correct the comparison table and the plugin-install walkthrough to reflect the real mcp__plugin_ruflo-core_ruflo__* tool names. Fix the `claude mcp add` example to use the server name actually registered (claude-flow). - catalog-manifest.json: regenerate (agents 164->165, current gitSha). - helpers/.helpers-version, helpers.manifest.json, statusline.cjs, .proven-config-version, proven-config.json: refresh to match current release metadata (forward version bump, not a revert). Co-Authored-By: RuFlo <ruv@ruv.net> Claude-Session: https://claude.ai/code/session_018ZkgSrVRWNMcXmaFiMZ7Z8 |
||
|
ruv
|
16b0dc4d5b |
chore(release): 3.38.16 → 3.38.17
Fixes #2990's root cause at the publish layer: v3/@claude-flow/cli and the claude-flow umbrella both pinned @claude-flow/mcp to 3.0.0-alpha.10, a version that was never published (npm registry tops out at alpha.9). Since @claude-flow/mcp is bundled (bundleDependencies) via a pnpm workspace link, this didn't affect the monorepo build, but it broke plain `npm install` against the published packages and left package.json/pnpm-lock.yaml out of sync with what's actually resolvable. Repinned to ^3.0.0-alpha.9 (the real latest, which already carries the protocolVersion-string fix) and regenerated v3/pnpm-lock.yaml accordingly. Co-Authored-By: RuFlo <ruv@ruv.net> Claude-Session: https://claude.ai/code/session_018ZkgSrVRWNMcXmaFiMZ7Z8 |
||
|
ruv
|
8a85001942 |
fix: Windows CI build, dead agentdb controller exports, memory driver doctor check
- #2992: pin windows-latest -> windows-2022 for the Build V3 matrix job. windows-latest moved to VS2026, which node-gyp can't detect, breaking native builds for hnswlib-node and better-sqlite3. - #2977: remove dead agentdb import attempts in controller-registry.ts for 8 exports no installable agentdb range provides. Promote the working tieredMemoryFallback/createConsolidationStub implementations to first-class for hierarchicalMemory/memoryConsolidation; the other 6 controllers (semanticRouter, mutationGuard, attestationLog, gnnService, rvfOptimizer, guardedVectorBackend) have no fallback and now return null directly instead of attempting an import that never resolves. The consolidation stub now reports source:'stub' and an explicit note so agentdb_consolidate reads as "did not run" rather than "nothing to do". - #2968: add a read-only doctor check (checkMemoryPersistenceDriver) that reports whether the active SQLite driver is native better-sqlite3 (durable, WAL-capable) or the sql.js fallback (silently drops wal_checkpoint writes), using the table-count signal from the issue (~47 native vs ~10 fallback). Warns, never fails; does not touch install behavior. #2990 (MCP HTTP transport protocolVersion/tools-list/IPv6 bind) required no code change — all three fixes are already present on main; the bug in published ruflo/@claude-flow/cli only persists because no release has been cut since they landed. #3002 (ruvocal hidden-tab stream freeze) also required no code change — already fixed on main. Co-Authored-By: RuFlo <ruv@ruv.net> Claude-Session: https://claude.ai/code/session_018ZkgSrVRWNMcXmaFiMZ7Z8 |
||
|
rUv
|
5234333c34 |
chore(release): 3.38.15 → 3.38.16 (#3081)
Cuts a release train off main after the 3 dream cycles landed unversioned: - #3062 dream(swarm): bound MessageBus retry attempts, fix unreachable message.failed (closes #3061) - #3057 dream(memory): hybridSearch controller reachable via explicit opt-in — was silently null-returning despite config flag (closes #3056) - #3049 dream(intelligence): discounted Thompson sampling for model-router bandit — opt-in prior decay recovers faster after workload shift (closes #3048) No code changes in this commit — just the version bump across all three publishable package.jsons. Claude-Session: https://claude.ai/code/session_0118jMsYhwHD5dx2vStENsEB |
||
|
rUv
|
b291ac4049 |
dream(intelligence): #3048 discounted Thompson sampling for model-router bandit (evaluated, ACCEPT-scoped) (#3049)
* dream(intelligence): discounted Thompson sampling for model-router bandit Adds an opt-in `priorDecay` config field (default 1 = disabled, fully backward compatible) to ModelRouter's Beta-Bernoulli bandit. When enabled, every bucket/model's alpha/beta decays geometrically once per recordOutcome() call before that round's reward is added (arXiv 2305.10718 discounted Thompson sampling), so accumulated routing history from a long-running persisted state file no longer permanently dominates the posterior after a real-world model-quality shift. Benchmark (n=30 paired trials, seeded PRNG, identical stream baseline vs candidate): non-stationary regime-shift recovery 26.5 -> 21.9 rounds (-17.6%, t=7.24); post-shift correct-routing rate +1.3pp (t=6.10); stationary-workload invariant held (delta +0.02pp, not a regression). Receipt: v3/@claude-flow/cli/benchmarks/results/prior-decay-receipt.json. Mirrors the epsilon-decay pattern q-learning-router.ts already applies to its own exploration rate -- an internal-consistency fix, not an imported pattern. Evaluation and adversarial critique in progress; this commit is the checkpointed, test-passing candidate + its baseline/candidate receipt. Co-Authored-By: RuFlo <ruv@ruv.net> * dream(intelligence): #3048 fix adversarial-critique findings, add gist + ledger Independent adversarial critique found and this commit fixes: a paired-t statistics bug (population vs sample stddev inflated every t-value by n/(n-1)), an unguarded priorDecay input (NaN/negative could silently poison persisted .swarm/model-router-state.json forever, bypassing sampleBeta's own defensive fallback), and a single-bucket benchmark blind spot (now covers low + med complexity buckets -- the med bucket shows no significant effect, disclosed rather than hidden). Also adds the research gist and backfills 3 missing ledger rows for 2026-08-14/15/16 (real PRs/issues existed for those nights but the ledger table wasn't updated) plus tonight's own row. Co-Authored-By: RuFlo <ruv@ruv.net> --------- Co-authored-by: Claude <noreply@anthropic.com> |
||
|
rUv
|
87bacd5582 |
chore(release): 3.38.14 → 3.38.15 (#3080)
Cuts a release train off main after the following PRs landed unversioned: - #3060 fix(cli): honor MCP tool selection when listing (closes #3055) - #3076 fix: resolve Claude Code launches on Windows (closes #3071) - #3077 fix(hooks): persist real session-end state (closes #3063) - #3044 dream(security): #3043 advisory scanner for untrusted settings.json hooks/allow-rules — CVE-2025-59536-class defense-in-depth, advisory-only (non-blocking), hardened against 6 concrete critic bypasses. No code changes in this commit — just the version bump across all three publishable package.jsons. Claude-Session: https://claude.ai/code/session_0118jMsYhwHD5dx2vStENsEB |
||
|
rUv
|
0e50fe7f57 |
dream(security): #3043 advisory scanner for untrusted settings.json hooks/allow-rules (evaluated, ACCEPT) (#3044)
* dream(security): #3043 advisory scanner for untrusted settings.json hooks/allow-rules carried forward by ruflo init/--upgrade mergeSettingsForUpgrade() and writeSettings() both read a target project's pre-existing .claude/settings.json and spread its hooks / permissions.allow entries into the merged output verbatim — the same trust shape as CVE-2025-59536 (settings.json hook payload achieving command execution with no review step), reached via ruflo init's own merge logic rather than Claude Code's loader. Neither of last night's flagged CVEs (2025-59536, 2025-6514) applies to Ruflo's own hook dispatch or OAuth client directly, but this distinct gap in the same family was genuinely unaddressed. Adds a static, advisory-only (non-blocking) scanner that inspects hook commands and Bash allow-rules for known-dangerous patterns before the merge and surfaces findings as CLI warnings. Merge/write behavior itself is completely unchanged. Hardened after an independent adversarial-critic pass found 6 concrete bypasses (intermediate-pipe-stage downloads, two-step download+exec, eval-wrapped base64, interpreter-mediated dangerous commands, absolute-path-prefixed commands, whitespace-evasive allow rules) — all fixed and pinned as a held-out regression set. Also fixes an ANSI/control-character injection risk in the warning output itself, self-identified during review. 87/87 relevant tests green (1 pre-existing environmental failure elsewhere in the package, unrelated — unbuilt sibling package). Dream Cycle 2026-08-16, security surface. Full research, evaluation receipts, and adversarial critique: issue #3043 and docs/dream-cycle/dream-gist-2026-08-16.md. Co-Authored-By: RuFlo <ruv@ruv.net> Claude-Session: https://claude.ai/code/session_01NVdeUqt73Bv7vnPxYpZy7i * docs(dream-cycle): update ledger — backfill 2026-08-14/2026-08-15, add 2026-08-16 2026-08-14 (swarm, REJECT) and 2026-08-15 (performance, REJECT) both had fully-run pipelines whose live ledger rows never reached main (every dream-cycle PR to date is unmerged, so branch-local ledger commits never merge forward) — verified via git ls-remote / search_pull_requests before backfilling, not assumed. 2026-08-16 (security, ACCEPT): advisory scanner for untrusted settings.json hooks/allow-rules carried forward by ruflo init/--upgrade. Issue #3043, PR #3044. Co-Authored-By: RuFlo <ruv@ruv.net> Claude-Session: https://claude.ai/code/session_01NVdeUqt73Bv7vnPxYpZy7i --------- Co-authored-by: Claude <noreply@anthropic.com> |
||
|
Wu Shuwen
|
02eae9c117 | fix(hooks): persist real session-end state (#3077) | ||
|
Wu Shuwen
|
f7aab68602 |
fix: resolve Claude Code launches on Windows (#3076)
* fix: launch Claude Code from Windows npm installs * test: use Windows path semantics in resolver |
||
|
Wu Shuwen
|
3a1b2257fb | fix(cli): honor MCP tool selection when listing (#3060) | ||
|
rUv
|
703e7741d7 |
fix(memory): #3051 memory_store tags lost via write-through cache (3.38.14) (#3079)
Root cause: bridgeStoreEntry's post-write cache set stored only a partial
entry shape — { id, key, namespace, content, embedding } — with no `tags`,
no timestamps, no metadata, no access_count. The DB row itself had the
tags JSON-serialized correctly. But any subsequent bridgeGetEntry hit the
cache first and returned `tags: cached.tags || []` = `[]`, so
`memory_store({ tags })` → `memory_retrieve()` looked like it silently
dropped tags.
Fix: replace the partial write-through with cache invalidation. The next
read then re-fetches the row via the DB reader (which correctly parses
the tags JSON column) and repopulates the cache from the full shape.
Cleaner than expanding the write-through payload — future entry-shape
fields don't need to remember to update the cache-write signature.
Regression test verifies: store {tags:[a,b,c]} → get returns {tags:[a,b,c]}.
Proven to catch the bug: pre-fix the test fails (tags: [] returned);
post-fix it passes.
The same defect would have caused reads to see access_count stuck at 0
and drifted createdAt/updatedAt timestamps from the moment the cache
was first populated until eviction; cache invalidation fixes those too.
Release
- Bumps @claude-flow/cli, claude-flow, ruflo: 3.38.13 → 3.38.14 (PATCH).
Claude-Session: https://claude.ai/code/session_0118jMsYhwHD5dx2vStENsEB
|
||
|
rUv
|
0f3c45101b |
fix: address #3045 #3064 #3065 + new ruflo-deepseek-harness plugin (3.38.13) (#3078)
Bug fixes
- #3045 statusline.cjs: getGitInfo() runs the 5-command git chain per render;
on large repos with concurrent Claude Code sessions this queued subprocesses
faster than they finished (reporter observed hundreds of orphan children +
load-avg in the hundreds). Added a per-cwd tmp file cache with 5s TTL —
dirty status still feels live, pileup is bounded.
- #3064 hooks post-task: the narrow ad-hoc regex /^[a-zA-Z0-9_-]+$/ silently
dropped every colon-namespaced plugin agent (ruflo-core:reviewer,
feature-dev:code-explorer, ...) — i.e. every Claude Code plugin agent. The
canonical validateIdentifier() upstream already allows ':' and '.', so the
redundant regex is removed. Regression test locks in 4 agent-shape cases,
proven to catch the bug: 2 pass / 2 fail on revert, 4 pass with fix.
- #3065 harness-gepa SKILL.md: unquoted colon in `description` broke YAML
parsing in `npx skills add`. Quoted the description; rephrased the bare
"(default:" to avoid the leading colon.
New plugin: ruflo-deepseek-harness
- plugins/ruflo-deepseek-harness/: sibling to ruflo-metaharness (ADR-150
shape). Two skills: `deepseek-chat` (non-reasoning) and `deepseek-reason`
(surfaces reasoning_content separately). Reads DEEPSEEK_API_KEY from env;
degrades gracefully (exit 0 with `{status: 'degraded', reason, hint}`)
when the key is missing or the API is unreachable. `--alert-on-error`
flag opts into hard exit 1 for CI gates. Smoke-tested locally.
Release
- Bump @claude-flow/cli, claude-flow, ruflo: 3.38.12 → 3.38.13 (PATCH:
bug fixes; the new plugin is scaffolding under plugins/ and not part of
the npm-published CLI packages).
Not fixed
- #3051 memory_store tags: current source's memory_store handler passes
tags straight through to storeEntry; a maintainer already verified live
round-trip works on
|
||
|
rUv
|
e43f8b9750 |
Merge pull request #3073 from ruvnet/fix/322c-receipt-roundtrip
fix(flywheel): compute receipt statistics from encoded scale-12 values — honest receipts no longer fail their own verifier (ruflo#3072) |
||
|
ruv
|
a38647e1f9 |
fix(flywheel): compute receipt statistics from encoded scale-12 values (ruflo#3072)
createFlywheelReceipt ran the bootstrap over full-precision score/delta doubles while the payload stored them as scale-12 decimal strings, and verifyFlywheelReceipt recomputes from those strings — the only values a verifier ever has. Any input that does not terminate within twelve decimals (e.g. a corpus mean of 0.7687083333333332, stored as "0.768708333333") shifted relativeLift by one ULP on re-decode, so an honest, correctly signed receipt failed its own verification with "statistical decision does not recompute" — indistinguishable from tampering. Encode baselineScore, candidateScore, heldOutDeltas, and frozenAnchorRegression first and feed the decoded encoded values into computePromotionStatistics, making producer and verifier bit-identical by construction. This implements the ADR-322C amendment in #3067: the statistical decision is computed from the encoded values of record. Regression test at the reproduction mean fails against the previous producer and passes with the fix. Co-Authored-By: claude-flow <ruv@ruv.net> Claude-Session: https://claude.ai/code/session_012Jib2gQyJpqCoo2xYAbb4X |
||
|
ruv
|
7d0bb9c53b |
fix(flywheel): reject unknown fields in receipt and commit verification
ADR-322C requires that unknown fields fail verification for a given schema version. verifyFlywheelReceipt did not enforce it: a signature is valid over whatever the producer canonicalized, so a producer-emitted field the contract never defined verified cleanly. Enforcement lived only in out-of-band JSON Schema validation, which is a step a consumer may or may not run, while verifyFlywheelReceipt is what every consumer calls. This matters now because rvm#35 and autogenous#10 are about to anchor promotion receipts into a witness chain and verify them across repo boundaries; a permissive verifier makes the anchor silently permissive too. Closes the class, not just the instance. verifyFlywheelLedger had the same weakness - a commit hashes consistently over whatever fields it carries, so an undefined field chained cleanly - and now runs the same check. Contract-open objects stay open, with a test pinning that: candidatePolicy is owned by policySchemaVersion, gates is a caller-named term map, and evidence.verification / evidence.canary carry evidence-specific payloads. Errors read "unknown field: <path>" so a caller can tell an unrecognized field from a tampered one, per the issue's triage requirement. Both regression tests were confirmed to fail with the checks disabled and pass with them enabled. Full flywheel suite: 37 passed (receipt 10, transaction 11, proposer 6, sequential-evidence 10). Strictness is versioned: a v1 verifier will reject a future receipt that adds fields. That is correct fail-closed behavior for a security property, and is now stated in ADR-322C's Update section (see #3069, PR #3067) so consumers plan for it rather than discover it. Refs #3068 Co-Authored-By: claude-flow <ruv@ruv.net> |
||
|
ruv
|
693385eab0 | test(ci): keep safety guard root-compatible | ||
|
ruv
|
13cbd697db | fix(release): enforce CI and idle assignment safety | ||
|
ruv
|
21f7c0adb8 | fix(release): align bundled runtime metadata | ||
|
ruv
|
392e71912e | fix(mcp): rate-limit authenticated HTTP routes | ||
|
ruv
|
2ec82b0cd1 | chore(release): prepare stable 3.38.10 train | ||
|
ruv
|
b9a5077292 | test(cli): model atomic fs operations in MCP suite |